Patient turning-over system based on intelligent perception
Through intelligent perception technology, multi-point pressure sensing data is used to analyze the changes in the patient's position and optimize the turnover movement, solving the problems of misjudgment and pressure ulcer risks in the turnover process in the existing technology, and achieving a more accurate and safe turnover process.
Patent Information
- Application Number
- CN202510444404.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-10
- Publication Date
- 2025-05-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the patient's turnover process, the existing bed relies on a single sensor data for position judgment, and lacks in-depth analysis of the trend of pressure change, which leads to the inability to accurately distinguish between active turnover behavior and unintentional movement, which can easily cause misjudgment, affect the rationality of turnover execution, and increase the risk of pressure ulcers in patients with long-term bed rest.
Using a patient turnover system based on intelligent perception, multi-point pressure sensing data is obtained through the pressure gradient sensing module, the pressure change rate and direction of adjacent sensing points are calculated, the initial trend of position adjustment is judged, and the active turnover and unintentional movement are distinguished. The position adjustment trend recognition module analyzes dynamic pressure gradient data and determines the target area for position adjustment. The turnover execution optimization module calculates the local pressure release rate, adjusts the turnover angle and execution speed. The local pressure regulation module monitors and optimizes the local pressure distribution to ensure balance of stress.
It realizes accurate judgment of the patient's position adjustment and optimization of turnover movements, improves the coordination and accuracy of the turnover process, reduces the pressure burden and pressure ulcers caused by improper turnover by patients with bed rest for a long time, and improves the quality of care and the comfort of the patient.
Smart Images
Figure CN119925104A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent sensing technology, and in particular to a patient turning system based on intelligent sensing. Background Art
[0002] The field of intelligent sensing technology includes technical methods for collecting, analyzing and processing environmental information, mainly involving the collaborative work of various sensors, data acquisition devices and computing units. The core content is to collect data from the physical world through sensors and use data processing units to analyze the information to support automated decision-making. Intelligent sensing technology is widely used in medical, industrial, transportation, environmental monitoring and other fields. Its main research directions include multi-sensor fusion, real-time data processing, intelligent identification and automatic control. The development of this technology has promoted the upgrading of many industries such as medical care, intelligent manufacturing and smart cities, and improved the accuracy of data acquisition and the level of intelligent decision-making.
[0003] Among them, the patient turning system based on intelligent perception refers to a system that uses sensors to obtain patient position information and performs turning operations in combination with the analysis results of the data processing unit. It targets health problems caused by the lack of autonomous movement in long-term bedridden patients, and covers technical matters such as information collection, data analysis, and actuator control. The specific methods include using pressure sensors and posture detection equipment to monitor the patient's position in real time, analyzing the trend of position changes through data calculation units, and combining electric drive devices (pulleys, etc.) to perform turning actions, while optimizing the turning time to reduce the risk of patients suffering from posture discomfort.
[0004] Existing beds rely on single sensor data to judge the patient's position during the patient's turning over, lacking in-depth analysis of the pressure change trend, resulting in the inability to accurately distinguish between active turning behavior and unintentional movement, which can easily lead to misjudgment and affect the rationality of turning execution. The processing method of pressure data fails to fully consider the cumulative gradient changes, resulting in deviations in the identification of the target area for patient position adjustment, which may cause improper adjustment of the turning direction. During the execution process, the existing technology lacks accurate calculation of the local pressure release rate during turning, resulting in an imbalance of force in the local area, and some pressure retention areas fail to be effectively released, increasing the risk of pressure sores in long-term bedridden patients. After turning over, the existing technology uses static pressure data for position adjustment, which cannot dynamically adapt to the natural changes in the patient's position, resulting in insufficient optimization of the support point, which can easily cause the patient to passively adjust his posture to relieve discomfort after turning over, making it difficult to achieve high-precision matching of the patient's position adjustment, affecting the nursing effect of long-term bedridden patients. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a patient turning system based on intelligent perception.
[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A patient turning system based on intelligent perception comprises: The pressure gradient sensing module obtains data from multiple pressure sensing nodes on the bed surface, calculates the pressure change rate of adjacent sensing points based on the time series, determines the direction of pressure change at each node, determines the initial trend of the patient's position adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data; The position adjustment trend identification module analyzes the directional distribution of the gradient change based on the dynamic pressure gradient data, determines the lateral pressure growth area of the bed, and determines the target area for the patient's position adjustment in combination with the pressure reduction in the adjacent area, and obtains the position adjustment trend analysis result; The turning execution optimization module calculates the local pressure release rate in each time period during the patient's turning process based on the position adjustment trend analysis result, compares the equilibrium threshold to determine the local force retention situation, adjusts the turning angle and execution speed according to the retention area, and obtains the turning adjustment optimization record; The local pressure control module monitors the pressure changes during the patient's turning over based on the turning adjustment optimization record, analyzes the pressure release state of the local area of the bed, compares the pressure distribution before and after turning over, adjusts the order of support points, and obtains the local pressure balanced distribution record.
[0007] As a further solution of the present invention, the dynamic pressure gradient data includes pressure change rate data of adjacent sensor nodes, pressure change direction records, and gradient cumulative change amount; the position adjustment trend analysis results include pressure growth areas, pressure reduction areas, and position adjustment target areas; the turning adjustment optimization records include local pressure release rate data, bed areas below the equilibrium threshold, local force retention conditions, turning angle adjustment records, and execution speed adjustment records; the local pressure balance distribution records include pressure release status, pressure distribution before and after turning, and support point adjustment sequence.
[0008] As a further solution of the present invention, the pressure gradient sensing module includes: The pressure data acquisition submodule obtains the data of multiple pressure sensor nodes on the bed surface, collects the pressure value of each sensor node, records the time series pressure data of adjacent sensor points, calculates the time series pressure change of each sensor node, and obtains the pressure change matrix; The gradient change calculation submodule calculates the pressure gradients of adjacent sensor nodes based on the pressure change matrix, and calculates the cumulative gradient changes between adjacent nodes according to the time series, using the formula: , Calculate the cumulative change in gradient , calculate the local gradient distribution according to the cumulative gradient change, and obtain the local gradient change data, where, represents the pressure value of the i-th sensor node at time t, represents the time interval between adjacent time points, represents the pressure value of a sensor node in another area at time t, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point. represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of another regional sensor node at time t-1; The posture adjustment trend determination submodule determines the pressure change direction of each sensor node based on the local gradient change data, counts the pressure gradient change trend of the sensor node, calculates the gradient change consistency of adjacent areas, determines the trend characteristics of posture adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data.
[0009] As a further solution of the present invention, the posture adjustment trend identification module includes: The gradient direction analysis submodule analyzes the gradient change direction of adjacent sensor nodes based on the dynamic pressure gradient data, calculates the direction stability in the time series, counts the overall distribution characteristics of the pressure gradient change, and obtains the gradient direction distribution data; The pressure area identification submodule calculates the pressure change trend on the side of the bed based on the gradient direction distribution data, determines the pressure increase area, and combines the pressure reduction in the adjacent area, using the formula: , Calculate the lateral pressure growth area value , analyze the pressure variation between regions and obtain the lateral pressure distribution data, where represents the pressure gradient of the ith node at time t, represents the pressure value of the jth sensor node at time t, n represents the number of sensor nodes, represents the pressure value of the jth sensor node at time t-1, represents the pressure gradient of the ith node at time t-1, represents the pressure gradient of the jth node at time t-1; The posture adjustment trend determination submodule analyzes the consistency of gradient changes in adjacent areas based on the lateral pressure distribution data, determines the target area for patient posture adjustment, and obtains posture adjustment trend analysis results in combination with directional trend changes.
[0010] As a further solution of the present invention, the turning execution optimization module includes: The turning over initial pressure extraction submodule extracts the pressure data of the initial stage of turning over based on the position adjustment trend analysis result, records the pressure changes in each time period during the turning over process, selects the local pressure value at the beginning of turning over, calculates the distribution of the initial force area, and obtains the turning over initial pressure distribution data; The local pressure release calculation submodule adopts the formula based on the initial pressure distribution data of turning over: , Calculating the local pressure release rate , analyze the relief trend of local pressure, screen the bed area below the equilibrium threshold, and obtain the local pressure release status record, among which, represents the pressure value of the i-th sensor node at time t, represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point; The retention area adjustment submodule determines the local force retention situation based on the local pressure release state record, selects the bed area where the continuous pressure is not released, adjusts the turning angle and execution speed for the retention area, and obtains the turning adjustment optimization record.
[0011] As a further solution of the present invention, the local pressure control module includes: The pressure monitoring submodule obtains the local pressure data at different time points during the patient's turning over according to the turning adjustment optimization record, monitors the pressure sensor data of multiple mattress support areas, records the pressure changes at various time points before and after turning over, calculates the pressure mean of each support area, and obtains the local pressure change data; The pressure distribution analysis submodule compares the pressure distribution of each support point before and after turning over based on the local pressure change data, using the formula: , Calculate the difference in local pressure distribution , and screen the area where the pressure is not released evenly to obtain the local pressure release deviation area, where and They represent the pressure values of the i-th support point before and after turning over, N is the total number of support points, and They represent the pressure values in a specific area j, respectively, and M is the number of support points in the area; The support point optimization adjustment submodule adjusts the release order of the support points in the area based on the local pressure release deviation area, adjusts the time difference of the mattress according to the pressure gradient sorting, and recalculates the adjusted pressure mean to obtain the local pressure balanced distribution record.
[0012] As a further solution of the present invention, the system further comprises a post-turnover support adjustment module; The support adjustment module after turning over collects the bed pressure data after turning over in real time based on the local pressure balance distribution record, calculates the pressure distribution under the current patient position, adjusts the mattress deformation and support points, and matches the new state of the patient's position.
[0013] As a further solution of the present invention, the post-turnover support adjustment module includes: The pressure data acquisition submodule obtains the pressure sensing data of the support area of the bed after turning over based on the local pressure balance distribution record, records the real-time pressure values of different support points, calculates the pressure average of each area, compares the pressure changes before and after turning over, analyzes the local pressure distribution of the support points, and obtains the pressure distribution data after turning over; The support area optimization submodule calculates the pressure change rate of the differential support area based on the pressure distribution data after turning over, selects the area with uneven pressure distribution, adjusts the deformation amplitude of the corresponding support point, adjusts the support strength of each area, balances the force state of different parts, and obtains the support point optimization adjustment plan; The patient position matching submodule adjusts the support point height and deformation amplitude according to the support point optimization adjustment plan, monitors the changes in the patient's position pressure distribution in real time, calculates the balance of the optimized support point, and matches the new state of the patient's position.
[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by acquiring multi-point pressure sensing data and calculating the pressure change rate of adjacent sensing points based on the time series, the initial trend of the patient's body position adjustment can be accurately judged, the pressure gradient direction distribution is analyzed, and the recognition ability of the target area of the body position adjustment is enhanced by combining the lateral pressure increase area of the bed and the pressure reduction of the adjacent area, and the trend prediction accuracy is improved. During the turning process, the calculation of the local pressure release rate ensures that the bed area below the equilibrium threshold can be identified, and combined with the local force retention situation, the turning angle and execution speed are dynamically adjusted to optimize the coordination of the turning action. During the turning process, the bed is continuously monitored. The pressure changes in local areas are analyzed through comparative analysis of the pressure distribution before and after turning over, and the order of adjusting the support points is optimized to balance the local force and reduce the risks brought by long-term local compression. After turning over, the surface pressure data of the bed is collected in real time, and the pressure distribution under the current patient position is calculated, the mattress deformation is adjusted and the support points are optimized, the body position matching is improved, and the discomfort after turning over is reduced. Through the refined control of the whole process of pressure perception, trend analysis, execution optimization, local regulation and support adjustment after turning over, the turning process is more accurate, which improves the comfort of bedridden patients, reduces the additional pressure burden caused by improper turning over, and improves the quality of nursing. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 is a system flow chart of the present invention; Figure 2 This is a flow chart of the pressure gradient sensing module of the present invention; Figure 3 This is a flow chart of the body position adjustment trend identification module of the present invention; Figure 4 It is a flow chart of the optimization module of the present invention; Figure 5 This is a flow chart of the local pressure control module of the present invention; Figure 6 This is a flow chart of the support adjustment module after turning over of the present invention. DETAILED DESCRIPTION
[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0018] See also Figure 1 , a patient turning system based on intelligent perception includes: The pressure gradient sensing module obtains data from multiple pressure sensing nodes on the bed surface, records the pressure value of each sensing node, and calculates the pressure change rate of adjacent sensing points based on the time series, determines the pressure change direction of each node, and counts the cumulative gradient changes of adjacent sensing nodes. It determines the initial trend of the patient's position adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data; The position adjustment trend recognition module analyzes the directional distribution of gradient changes based on dynamic pressure gradient data, determines the pressure increase area on the side of the bed, and determines the target area for patient position adjustment based on the pressure reduction in the adjacent area, and obtains the position adjustment trend analysis results; The turning execution optimization module extracts the pressure data in the initial stage of turning based on the trend analysis results of body position adjustment, calculates the local pressure release rate in each time period during the patient's turning process, screens the bed area below the equilibrium threshold, determines the local force retention, adjusts the turning angle and execution speed for the retained area, and obtains the turning adjustment optimization record; The local pressure control module monitors the pressure changes during the patient's turning process based on the turning adjustment optimization record, analyzes the pressure release status of the local area of the bed, compares the pressure distribution before and after turning, adjusts the order of support points to make the local force uniform, and obtains the local pressure balance distribution record; The support adjustment module after turning over is based on the local pressure balance distribution record, collects the bed pressure data after turning over in real time, calculates the pressure distribution under the current patient position, adjusts the mattress deformation and optimizes the support points to match the new state of the patient's position.
[0019] Dynamic pressure gradient data include pressure change rate data of adjacent sensor nodes, pressure change direction records, and gradient cumulative change; position adjustment trend analysis results include pressure growth area, pressure reduction area, and position adjustment target area; turning adjustment optimization records include local pressure release rate data, bed area below the equilibrium threshold, local force retention, turning angle adjustment records, and execution speed adjustment records; local pressure balance distribution records include pressure release status, pressure distribution before and after turning, and support point adjustment sequence.
[0020] See also Figure 2 , the pressure gradient sensing module includes: The pressure data acquisition submodule obtains the data of multiple pressure sensor nodes on the bed surface, collects the pressure value of each sensor node, records the time series pressure data of adjacent sensor points, calculates the time series pressure change of each sensor node, and obtains the pressure change matrix; Obtain the data of multi-point pressure sensing nodes on the bed surface, select n sensors within the coverage of the bed surface, and record the initial pressure value for subsequent calculation; the pressure value of each sensing node is collected by a high-precision pressure sensor. The distribution density of the sensor depends on the size of the detection area. For example, 25-50 sensors can be arranged within 1m², and the detection range of each sensor is about 5cm×5cm, which can ensure fine-grained pressure collection; in the time dimension, set the sampling interval Δt=0.1s, and record all pressure data within T=60s to form a time series data set; the pressure value of adjacent sensing points is collected by high-precision pressure sensors. The distribution density of the sensors depends on the size of the detection area. For example, 25-50 sensors can be arranged within 1m², and the detection range of each sensor is about 5cm×5cm, which can ensure fine-grained pressure collection; in the time dimension, set the sampling interval Δt=0.1s, and record all pressure data within T=60s to form a time series data set; The pressure data are stored in the matrix P(t, b) in chronological order, where t represents the time index and b represents the sensor point number. For example, at t=5s, the pressure values of sensor points 1, 2, and 3 can be expressed as P(5, 1), P(5, 2), P(5, 3), and so on, and are extended to all sensor points. Based on the time series data, the pressure change ΔP of the sensor point within the adjacent time step Δt is calculated, which is defined as ΔP=P(t)-P(t-1), that is, at t=5s, the pressure change of sensor point 1 is ΔP(5, 1)=P(5, 1)-P(4, 1).
[0021] The pressure variation matrix is shown in Table 1.1 below: Table 1.1 Pressure variation matrix Time (s) Sensing point 1 (Pa) Sensing point 2 (Pa) Sensing point 3 (Pa) Sensing point 4 (Pa) Sensing point 5 (Pa) 1 0 0 0 0 0 2 1.2 0.8 1.0 1.1 0.9 3 1.5 1.1 1.3 1.4 1.2 4 1.7 1.4 1.5 1.6 1.5 5 2.0 1.8 1.7 1.9 1.8 As shown in Table 1.1, the time series data stores the pressure changes of multiple sensor nodes at different time points. The data in each cell represents the pressure change value of the corresponding sensor point. Through the matrix data, the cumulative change of the pressure gradient can be further calculated, and finally the pressure change matrix can be obtained.
[0022] The gradient change calculation submodule calculates the pressure gradients of adjacent sensor nodes based on the pressure change matrix, and calculates the cumulative gradient changes between adjacent nodes according to the time series, using the formula: , Calculate the cumulative change in gradient , calculate the local gradient distribution according to the cumulative gradient change, and obtain the local gradient change data, where, represents the pressure value of the i-th sensor node at time t, represents the time interval between adjacent time points, represents the pressure value of a sensor node in another area at time t, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point. represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of another regional sensor node at time t-1; Based on the pressure change matrix, the pressure gradients of adjacent sensor nodes are calculated. The gradient change of each sensor point is determined by the pressure change of itself and its adjacent nodes. For example, if the pressure values of adjacent sensor points i and j are P(i, t) and P(j, t) respectively, is 0.1s, that is, the gradient change is calculated every 0.1s, and m represents the number of neighborhood sensors of a sensor point. For example, in a 5x5 grid, the number of neighbors of the central point is 8. At t=5s, select a sensor point i=3 and calculate its gradient change: The pressure value at this point is P (5, 3) = 1.7 Pa; The pressure values of adjacent points (numbers 2 and 4) are P (5, 2) = 1.8 Pa and P (5, 4) = 1.9 Pa respectively; Time step ; Substitute into the formula to calculate the gradient changes: ; ; ; The calculated cumulative gradient change = 4.78. Through this calculation, the local gradient distribution of all sensor nodes can be obtained, and finally the local gradient change data can be obtained.
[0023] The posture adjustment trend determination submodule determines the pressure change direction of each sensor node based on the local gradient change data, counts the pressure gradient change trend of the sensor node, calculates the gradient change consistency of adjacent areas, identifies the trend characteristics of posture adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data; Based on the local gradient change data, the pressure change direction of each sensor node is judged. By calculating the gradient change trend of the time series data and counting the gradient change consistency of adjacent areas, the judgment formula is defined: ; in, represents the trend of body position adjustment, q represents the number of time windows for continuous calculation. For example, if q=5 is set, the trend is calculated for the data of the last five time segments. If the gradient change of a certain sensor point at t=5s is 4.78, and the gradient changes of the first four time points are 4.0, 4.1, 4.3, and 4.5 respectively, then the calculation is: ; ; ; The calculated trend value of body position adjustment is 7.8. Combined with the time series data, the basis for determining this value comes from clinical data analysis. The pressure change data of 1,000 patients are statistically analyzed, and the distribution interval of trend values under active turning and unintentional movement behaviors is extracted. The trend value distribution interval is calibrated according to the behavioral records of the subjects. The statistical analysis shows that: When the subject turns over actively, The value is usually between 6.0-10.0, accounting for 85%, and the corresponding turning angle change is greater than 30°, accompanied by an increase in the synchronization of the pressure change rate in multiple regions; When the subject does not intend to move, The value is usually lower than 6.0, accounting for 90%. At this time, the local pressure fluctuation amplitude is small, and the pressure change rate does not have a synchronous trend; like >10.0 indicates that the subject has large displacement behaviors, such as sitting up or violent movements, which accounts for less than 5%. Such situations need to be monitored and handled separately.
[0024] The trend value range of 6.0-10.0 is set based on: By analyzing the cumulative amount of gradient changes in adjacent sensor areas, it was found that when the turning angle was >30°, the corresponding gradient fluctuation amplitude was much higher than the local unintentional movement behavior; During the turning process, the pressure curves of multiple sensing points show a continuous change trend, and the deviation trajectory of the pressure center point is continuous and stable; Based on the subjects' motor behavior data, the turning-over time window was set to 2-3 seconds, and when the trend value was between 6.0 and 10.0, it corresponded to a 90% probability of active turning-over behavior; Finally, by combining the statistical analysis and calculation process, it was determined that if the trend value was in the range of 6.0-10.0, it was judged as active turning behavior, and if the trend value was lower than 6.0, it was judged as unintentional movement, and finally the dynamic pressure gradient data was obtained.
[0025] See also Figure 3 , the posture adjustment trend recognition module includes: The gradient direction analysis submodule analyzes the gradient change direction of adjacent sensor nodes based on dynamic pressure gradient data, calculates the directional stability in the time series, counts the overall distribution characteristics of the pressure gradient change, and obtains the gradient direction distribution data; Based on the dynamic pressure gradient data on the bed surface, the gradient change direction of adjacent sensor nodes is analyzed. First, pressure data needs to be collected from multiple sensor nodes to form a time series pressure matrix. For each sensor node The pressure at time t , calculate the gradient change between time t-1 and t ,like , the pressure at the node increases, indicating that the area may be further contacted by the patient's body. , it indicates that the pressure in the area is reduced and the body position may change. In the data acquisition process, assuming that the sensor network consists of 5 sensor nodes, the pressure data at a certain moment is as follows: Table 2.1 Initial pressure table of monitoring points
[0026] According to the monitoring data in Table 2.1, the gradient change can be calculated : ; ; ; ; ; By marking the pressure change trend of each sensing point on the gridded bed surface, the gradient change direction distribution can be obtained and the gradient direction distribution data can be generated.
[0027] The pressure area identification submodule calculates the pressure change trend on the side of the bed based on the gradient direction distribution data, determines the pressure increase area, and combines the pressure reduction in the adjacent area, using the formula: , Calculate the lateral pressure growth area value , analyze the pressure variation between regions and obtain the lateral pressure distribution data, where represents the pressure gradient of the ith node at time t, represents the pressure value of the jth sensor node at time t, n represents the number of sensor nodes, represents the pressure value of the jth sensor node at time t-1, represents the pressure gradient of the ith node at time t-1, represents the pressure gradient of the jth node at time t-1; Based on the gradient direction distribution data, the pressure change trend on the side of the bed is analyzed. First, the pressure change of the adjacent areas is calculated. If the pressure increase in a certain area is accompanied by a decrease in the pressure in the adjacent area, it can be determined that the area may be the target area for position adjustment. The formula is used to calculate the lateral pressure growth area value.
[0028] Assume that a bed area monitors 5 nodes in total, and the gradient changes of some sensor nodes are as follows: Table 2.2 Sensor gradient change table
[0029] calculate: ; ; ; Operation results: ; If the value is greater than 0, it means that there is a significant pressure change in the current bed area, and the lateral pressure distribution data is obtained.
[0030] The position adjustment trend determination submodule analyzes the consistency of gradient changes in adjacent areas based on the lateral pressure distribution data, determines the target area for the patient's position adjustment, and obtains the position adjustment trend analysis results in combination with the directional trend changes; Based on the lateral pressure distribution data, the consistency of gradient changes in adjacent areas is analyzed to determine the target area for patient position adjustment. First, calculate whether the gradient changes in adjacent areas meet specific trend requirements, that is, whether there is an increase in pressure in area A and a decrease in pressure in area B, and whether they meet the consistency in time series. Set the threshold for determining the pressure change trend This value must ensure that it can effectively distinguish between the patient's position adjustment behavior and unintentional movement, so its basis is as follows: Based on the characteristics of human body pressure distribution changes: In clinical monitoring, there is a stable range of changes in the pressure area under different body positions. Under normal circumstances, a complete body position adjustment will cause the pressure change rate between adjacent areas to exceed 30%, that is, the ratio of the pressure increase or decrease of adjacent sensor nodes should exceed 0.3 times the pressure at the previous moment; Combined with actual bed sensor data: In the collected bed sensor data, every time a long-term bedridden patient turns over, the pressure fluctuation range of the adjacent area is usually between 0.4-0.6, and in a few cases it will exceed this range; Calculate the pressure change distribution: If the pressure increase in a certain area is significant, then its gradient change Gradient changes with adjacent areas There is a negative correlation, that is hour, ,and Certain thresholds should be met; Based on the above factors, set the pressure change trend judgment threshold , when the calculated lateral pressure growth area value When it is greater than the threshold, it is determined that a posture adjustment trend occurs.
[0031] In the above calculation, we have obtained =0.59, due to , the judgment conditions are met, so the patient's position adjustment trend can be determined, and the adjustment target area can be marked to obtain the position adjustment trend analysis results.
[0032] See also Figure 4 , the turnaround execution optimization module includes: The turning initial pressure extraction submodule extracts the pressure data of the initial stage of turning based on the position adjustment trend analysis results, records the pressure changes in each time period during the turning process, selects the local pressure value at the beginning of turning, calculates the distribution of the initial force area, and obtains the initial pressure distribution data of turning; Based on the results of the trend analysis of body position adjustment, the pressure data of the initial stage of turning over are extracted to monitor the pressure distribution of the patient in the initial stage of turning over. The multi-point sensors on the surface of the bed record the pressure values of each contact point and form a pressure time series matrix, in which the matrix elements under each time frame represent the instantaneous pressure value of the corresponding point. For example, At this moment, the pressure in the head area is 45 mmHg, the back area is 80 mmHg, and the hip area is 120 mmHg. Then, as the body turns over, the pressure values at each point change. These data can constitute the initial pressure distribution matrix.
[0033] In the process of extracting data, it is necessary to normalize the pressure changes in multiple time frames and calculate the changes in pressure values per unit time. If the pressure change in a certain area is greater than 10 mmHg in three consecutive time frames, it can be determined that the area is undergoing a turning force adjustment, otherwise it is still in a static state. arrive During this time, the pressure in the buttocks area dropped from 120mmHg to 95mmHg, and the pressure in the back area dropped from 80mmHg to 50mmHg. Combined with the stable pressure of 45mmHg in the head area, it can be preliminarily judged that the patient's turning over is in progress, and the main force-bearing areas are the buttocks and back.
[0034] In order to clarify the force state in the initial stage of turning over, it is necessary to further calculate the local force balance and set the force balance threshold. =60mmHg. The value is set based on the following: Medical research shows that prolonged pressure on local tissues above 32mmHg (capillary closure pressure) may lead to local blood supply obstruction, while prolonged pressure above 60mmHg will significantly increase the risk of pressure sores. =60mmHg is used as the criterion for judging force balance during turning over. That is, when the pressure in a certain area is lower than 60mmHg, it means that the local blood supply has been restored and the turning adjustment effect is good; when it is higher than 60mmHg, retained pressure may still exist and further adjustment is required.
[0035] For example, in the initial stage of turning over, if the pressure of the back area is 50 mmHg, which is lower than the threshold, and the pressure of the buttocks area is 95 mmHg, which is higher than the threshold, further attention should be paid to the buttocks area to determine its pressure release situation.
[0036] Combining the above process, the initial pressure distribution data of turning over is finally obtained, which includes the initial pressure value of each area, the balance judgment result and the force area information of the turning over action. As shown in Table 3.1, the pressure data of each area of a patient in the initial stage of turning over is recorded.
[0037] Table 3.1 Initial pressure distribution data for turning over Time t(s) Head pressure (mmHg) Back pressure (mmHg) Buttocks pressure (mmHg) <![CDATA[t0]]> 45 80 120 <![CDATA[t1]]> 45 70 110 <![CDATA[t2]]> 45 50 95 As shown in Table 3.1, the pressure in the back area gradually decreased and tended to be balanced over time, while the pressure in the buttocks remained at a high level and its release needed to be further evaluated in subsequent steps.
[0038] The local pressure release calculation submodule is based on the initial pressure distribution data of turning over, using the formula: , Calculating the local pressure release rate , analyze the relief trend of local pressure, screen the bed area below the equilibrium threshold, and obtain the local pressure release status record, among which, represents the pressure value of the i-th sensor node at time t, represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point; Based on the initial pressure distribution data of turning over, the local pressure release rate at different time periods during the patient's turning over is calculated to determine whether there is local force retention. In this process, the pressure data of the buttocks, back and head are selected for calculation, and the time interval is set. seconds, and the pressure release rate of each area is calculated using the formula.
[0039] Assuming n=3 (representing the three areas of head, back and buttocks), substitute the data in Table 3.1 to calculate the local pressure release rate.
[0040] Calculate the numerator part (sum of squared changes): ; Calculate the denominator (the sum of the absolute changes): ; Calculate the square root of the mean: ; ; The final calculated pressure release rate is: ; Calculate the local pressure release rate , further evaluation of the retention area is required.
[0041] The retention area adjustment submodule determines the local force retention situation based on the local pressure release status record, selects the bed area where the continuous pressure is not released, adjusts the turning angle and execution speed for the retention area, and obtains the turning adjustment optimization record; Based on the local pressure release status, the local force retention situation is judged, and the bed area where the continuous pressure is not released is screened. Assuming the balanced pressure threshold =20mmHg. If the pressure in a certain area is lower than this value, it is considered that the force is balanced. If it is higher than this value and the pressure release rate is lower than =10mmHg / s, there is a risk of pressure retention in the area.
[0042] =10mmHg / s is set based on the following: According to the clinical nursing standards for turning over, the pressure release rate during normal turning over should usually be between 8-15mmHg / s. If the release rate is lower than 8mmHg / s, it means that the turning over adjustment is insufficient, which may lead to local pressure retention and increase the risk of tissue damage. If it is higher than 15mmHg / s, it may mean that the turning angle or execution speed is too large, which may easily cause discomfort to the patient. Therefore, the setting =10mmHg / s is used as the lower limit of the pressure release rate to ensure that the pressure relief rate during turning over is moderate.
[0043] According to the above calculation, the pressure in the hip area is 95 mmHg higher than =20 mmHg, and , indicating that the current pressure release rate is acceptable, but if the release rate is lower than the threshold, the turning angle needs to be adjusted. Assuming the current turning angle is , execute the turning angle adjustment formula: ; in, The adjustment coefficient is set based on the following: a 1° increase in the turning angle can release about 10 mmHg of local pressure, so the adjustment coefficient is set to 0.1° / mmHg. Substitute into the calculation: ; After calculation, the adjustment angle is 23.5°. After adjustment, the pressure change should be observed. If the release rate is still lower than the threshold, the turning speed needs to be further adjusted, for example, from the current Adjust to .
[0044] Finally, based on the above adjustments, an optimized record of turning adjustment is obtained, which includes the adjusted turning angle, execution speed, and determination results of the pressure retention state, to ensure that the pressure is evenly released during the patient's turning process.
[0045] See also Figure 5 , the local pressure control module includes: The pressure monitoring submodule adjusts and optimizes the records according to the turning process, obtains the local pressure data at different time points during the patient's turning process, monitors the pressure sensor data of multiple mattress support areas, records the pressure changes at various time points before and after turning, calculates the mean pressure of each support area, and obtains the local pressure change data; During the patient's turning over, local pressure data at different time points are obtained. First, high-precision pressure sensors are arranged in each support area of the mattress. These sensors can detect and record the pressure value of each area in real time. Assume that the mattress is divided into six support areas, namely A, B, C, D, E and F. A pressure sensor is installed in each area. Before the patient turns over, the initial pressure value of each area is recorded. Assume that the following data is obtained: Table 4.1 Initial pressure table of monitoring points area Initial pressure value (mmHg) A 30 B 45 C 50 D 40 E 35 F 25 As shown in Table 4.1, during the turning process, the data of these sensors are continuously monitored, and the pressure changes after turning are recorded. Assume that the following data are obtained after turning: Table 4.2 Pressure gauge after monitoring point turning over area Pressure value after turning over (mmHg) A 25 B 40 C 45 D 35 E 30 F 20 Calculate the average pressure of each support area. The average pressure of each area before turning over is calculated as follows: ; The average pressure of each area after turning over is calculated as follows: ; Through the above calculation, local pressure change data is obtained.
[0046] The pressure distribution analysis submodule compares the pressure distribution of each support point before and after turning over based on the local pressure change data, using the formula: , Calculate the difference in local pressure distribution , and screen the area where the pressure is not released evenly to obtain the local pressure release deviation area, where and They represent the pressure values of the i-th support point before and after turning over, N is the total number of support points, and They represent the pressure values in a specific area j, respectively, and M is the number of support points in the area; Based on the local pressure change data, the difference in pressure release in each support area is calculated. First, the pressure difference before and after turning over in each area is calculated, as follows: Table 4.3 Pressure difference table of monitoring points area Pressure difference (mmHg) A 30-25=5 B 45-40=5 C 50-45=5 D 40-35=5 E 35-30=5 F 25-20=5 Calculate the local pressure distribution difference, where n is the total number of support points, which is 6, and m is the number of support points in a specific area, which is assumed to be 2. The calculation is as follows: ; ; According to the difference in local pressure distribution, areas where pressure is not released evenly are screened. According to the pressure balance threshold of 20 mmHg set above, if the pressure difference in a certain area is greater than this threshold, it is considered that the pressure release in this area is uneven. However, in this example, the pressure difference in all areas is 5 mmHg, which is less than 20 mmHg. Therefore, no area with uneven pressure release is detected. Finally, the local pressure release deviation area is obtained.
[0047] The support point optimization and adjustment submodule adjusts the release order of the support points in the local pressure release deviation area based on the local pressure release deviation area, adjusts the time difference of the mattress according to the pressure gradient sorting, and recalculates the adjusted pressure mean to obtain the local pressure balance distribution record; Call the local pressure release deviation area to adjust the release order of the support points in the area. Suppose in actual application, the pressure release in some areas is uneven, and the support points in these areas need to be adjusted. Specifically, according to the pressure gradient sorting, adjust the time difference between the inflation and deflation of the mattress airbags, so that the airbags at the high-pressure points are deflated first, and the airbags at the low-pressure points are deflated later. Suppose there are three support points in a certain area, and their initial pressure values are 60mmHg, 50mmHg and 40mmHg respectively. In order to achieve pressure balance, first sort the pressure of these three support points, and then set the time difference between the inflation and deflation of the airbags according to the sorting results, so that the airbags at the high-pressure points are deflated first, and the airbags at the low-pressure points are deflated later. After adjustment, recalculate the average pressure of each support point. Suppose the adjusted pressure value is as follows: Table 4.4 Support point pressure table after adjustment Support Points Adjusted pressure value (mmHg) 1 45 2 45 3 45 At this time, the average pressure of each support point is: , Through the above adjustments, the purpose of pressure balance at each supporting point is achieved, and finally, the local pressure balance distribution record is obtained.
[0048] See also Figure 6 , the support adjustment module after turning over includes: The pressure data acquisition submodule obtains the pressure sensing data of the bed support area after turning over based on the local pressure balance distribution record, records the real-time pressure values of different support points, calculates the pressure average of each area, compares the pressure changes before and after turning over, analyzes the local pressure distribution of the support points, and obtains the pressure distribution data after turning over; After the bed is turned over, pressure sensors are first arranged to monitor each support area of the bed. The sampling frequency of the sensor is set to 10 Hz to ensure the continuity and accuracy of the data. The collected data includes timestamp, pressure value, support point coordinates, etc. The pressure data of each support point is arranged in time series to facilitate the subsequent calculation of the pressure change value at different time points. For example, after turning over, the pressure of a certain support point is recorded as 320Pa, 310Pa, and 305Pa at the 1st, 3rd, and 5th seconds, respectively, indicating that the pressure is gradually decreasing. In order to calculate the mean pressure of each support area, the pressure data of adjacent areas are selected for averaging to reduce the abnormal influence of single-point data. Assuming that an area contains 4 support points, and their pressure values are 300Pa, 320Pa, 310Pa, and 290Pa, respectively, the mean pressure of the area is calculated as follows: ; After calculating the mean pressure of each area, it is necessary to further compare the pressure changes before and after turning over. Assuming that the mean pressure of the area before turning over is 330Pa, the pressure change value is calculated: ; This calculation method can be applied to all support areas, and the pressure distribution data after turning over can be obtained through batch calculation.
[0049] The support area optimization submodule calculates the pressure change rate of the differential support area based on the pressure distribution data after turning over, selects the area with uneven pressure distribution, adjusts the deformation amplitude of the corresponding support point, adjusts the support strength of each area, balances the force state of different parts, and obtains the support point optimization adjustment plan; After calculating the pressure distribution data after turning over, it is necessary to further calculate the pressure change rate of different support areas to evaluate the local pressure change. The calculation formula is as follows: ; in, is the pressure change value, is the pressure value before turning over. Assuming that the pressure of a certain area before turning over is 350Pa, and the pressure drops to 300Pa after turning over, the pressure change rate is: ; According to experience, if the pressure change rate exceeds 10%, it is determined that the deformation amplitude of the support point in the area needs to be adjusted. The specific adjustment methods include increasing or decreasing the airbag pressure in the area or adjusting the height of the mechanical support unit to optimize the mattress support structure. The pressure of each area after adjustment needs to be calculated twice to ensure that the adjusted balance meets the set standard. Assuming that the average pressure of a certain area becomes 315Pa after adjustment, the new pressure change rate is calculated: ; If the threshold value is within 10%, the adjustment of the area is completed, otherwise the optimization and adjustment will continue, and finally the support point optimization and adjustment plan will be obtained.
[0050] The patient position matching submodule adjusts the support point height and deformation amplitude according to the support point optimization adjustment plan, monitors the changes in the patient's position pressure distribution in real time, calculates the balance of the optimized support point, and matches the new state of the patient's position; After the support point optimization adjustment plan is completed, the deformation amplitude of the adjusted support point needs to be adjusted, and the pressure distribution changes of the patient's body position need to be monitored. Assuming that a patient weighs 80kg, the average pressure of the trunk support area after turning over is 320Pa, but the pressure in the leg area is lower, only 260Pa. At this time, the support point of the leg area needs to be adjusted to increase its pressure to close to the pressure of the trunk support area to achieve balanced matching. In order to achieve this adjustment, the inflation pressure of the leg support area can be increased. For example, if the original airbag pressure is 20kPa, it can be increased to 22kPa. After adjustment, recalculate the pressure distribution balance: ; in, is the mean pressure in the trunk area, is the average pressure in the leg area. After adjustment, the pressures are 320Pa and 310Pa respectively. The balance degree is calculated as follows: ; If the balance is lower than 5%, the support matching is considered complete, otherwise the optimization and adjustment will continue to be carried out to finally obtain the matching data of the patient's new body position.
[0051] Data Table Table 5.1 Pressure values of different areas before and after turning over Support area Pressure before turning over (Pa) Pressure after turning over (Pa) Pressure change (Pa) Pressure change rate (%) head 250 240 10 4.0 Chest 320 290 30 9.4 abdomen 350 300 50 14.3 Hips 400 370 30 7.5 Legs 270 260 10 3.7 As shown in Table 5.1, there are differences in the pressure distribution in each area after turning over, among which the pressure change rate in the abdominal area reaches 14.3%, which exceeds the set threshold of 10% and needs to be optimized and adjusted.
[0052] Table 5.2 Pressure balance after support point optimization and adjustment Adjust the front leg pressure (Pa) Adjust the leg pressure (Pa) Trunk pressure (Pa) Pressure balance (%) 260 310 320 3.1 As shown in Table 5.2, after the support point optimization and adjustment, the pressure in the leg area increased from 260Pa to 310Pa, and the balance with the trunk pressure of 320Pa was reduced to 3.1%, meeting the optimization goal.
[0053] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. A patient turning system based on intelligent perception, characterized in that: The system comprises: The pressure gradient sensing module obtains data from multiple pressure sensing nodes on the bed surface, calculates the pressure change rate of adjacent sensing points based on the time series, determines the direction of pressure change at each node, determines the initial trend of the patient's position adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data; The position adjustment trend identification module analyzes the directional distribution of the gradient change based on the dynamic pressure gradient data, determines the lateral pressure growth area of the bed, and determines the target area for the patient's position adjustment in combination with the pressure reduction in the adjacent area, and obtains the position adjustment trend analysis result; The turning execution optimization module calculates the local pressure release rate in each time period during the patient's turning process based on the position adjustment trend analysis result, compares the equilibrium threshold to determine the local force retention situation, adjusts the turning angle and execution speed according to the retention area, and obtains the turning adjustment optimization record; The local pressure control module monitors the pressure changes during the patient's turning over based on the turning adjustment optimization record, analyzes the pressure release state of the local area of the bed, compares the pressure distribution before and after turning over, adjusts the order of support points, and obtains the local pressure balanced distribution record.
2. The patient turning system based on intelligent perception according to claim 1 is characterized in that: The dynamic pressure gradient data includes pressure change rate data of adjacent sensor nodes, pressure change direction records, and gradient cumulative change amount; the posture adjustment trend analysis results include pressure growth areas, pressure reduction areas, and posture adjustment target areas; the turning adjustment optimization records include local pressure release rate data, bed areas below the equilibrium threshold, local force retention conditions, turning angle adjustment records, and execution speed adjustment records; the local pressure equilibrium distribution records include pressure release status, pressure distribution before and after turning, and support point adjustment sequence.
3. The patient turning system based on intelligent perception according to claim 1, characterized in that: The pressure gradient sensing module comprises: The pressure data acquisition submodule obtains the data of multiple pressure sensor nodes on the bed surface, collects the pressure value of each sensor node, records the time series pressure data of adjacent sensor points, calculates the time series pressure change of each sensor node, and obtains the pressure change matrix; The gradient change calculation submodule calculates the pressure gradients of adjacent sensor nodes based on the pressure change matrix, and calculates the cumulative gradient changes between adjacent nodes according to the time series, using the formula: ， Calculate the cumulative change in gradient , calculate the local gradient distribution according to the cumulative gradient change, and obtain the local gradient change data, where, represents the pressure value of the i-th sensor node at time t, represents the time interval between adjacent time points, represents the pressure value of a sensor node in another area at time t, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point. represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of another regional sensor node at time t-1; The posture adjustment trend determination submodule determines the pressure change direction of each sensor node based on the local gradient change data, counts the pressure gradient change trend of the sensor node, calculates the gradient change consistency of adjacent areas, determines the trend characteristics of posture adjustment, distinguishes active turning behavior from unintentional movement, and obtains dynamic pressure gradient data.
4. The patient turning system based on intelligent perception according to claim 1 is characterized in that: The posture adjustment trend identification module includes: The gradient direction analysis submodule analyzes the gradient change direction of adjacent sensor nodes based on the dynamic pressure gradient data, calculates the direction stability in the time series, counts the overall distribution characteristics of the pressure gradient change, and obtains the gradient direction distribution data; The pressure area identification submodule calculates the pressure change trend on the side of the bed based on the gradient direction distribution data, determines the pressure increase area, and combines the pressure reduction in the adjacent area, using the formula: ， Calculate the lateral pressure growth area value , analyze the pressure variation between regions and obtain the lateral pressure distribution data, where represents the pressure gradient of the ith node at time t, represents the pressure value of the jth sensor node at time t, n represents the number of sensor nodes, represents the pressure value of the jth sensor node at time t-1, represents the pressure gradient of the ith node at time t-1, represents the pressure gradient of the jth node at time t-1; The posture adjustment trend determination submodule analyzes the consistency of gradient changes in adjacent areas based on the lateral pressure distribution data, determines the target area for patient posture adjustment, and obtains posture adjustment trend analysis results in combination with directional trend changes.
5. The patient turning system based on intelligent perception according to claim 1, characterized in that: The turning over execution optimization module comprises: The turning over initial pressure extraction submodule extracts the pressure data of the initial stage of turning over based on the position adjustment trend analysis result, records the pressure changes in each time period during the turning over process, selects the local pressure value at the beginning of turning over, calculates the distribution of the initial force area, and obtains the turning over initial pressure distribution data; The local pressure release calculation submodule adopts the formula based on the initial pressure distribution data of turning over: ， Calculating the local pressure release rate , analyze the relief trend of local pressure, screen the bed area below the equilibrium threshold, and obtain the local pressure release status record, among which, represents the pressure value of the i-th sensor node at time t, represents the pressure value of the i-th sensor node at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, represents the pressure value of the adjacent j sensor nodes at time t, represents the pressure value of the adjacent j sensor nodes at time t-1, n represents the number of sensor nodes, and m represents the number of sensors in the neighborhood of the sensor point; The retention area adjustment submodule determines the local force retention situation based on the local pressure release state record, selects the bed area where the continuous pressure is not released, adjusts the turning angle and execution speed for the retention area, and obtains the turning adjustment optimization record.
6. The patient turning system based on intelligent perception according to claim 1, characterized in that: The local pressure control module comprises: The pressure monitoring submodule obtains the local pressure data at different time points during the patient's turning over according to the turning adjustment optimization record, monitors the pressure sensor data of multiple mattress support areas, records the pressure changes at various time points before and after turning over, calculates the pressure mean of each support area, and obtains the local pressure change data; The pressure distribution analysis submodule compares the pressure distribution of each support point before and after turning over based on the local pressure change data, using the formula: ， Calculate the difference in local pressure distribution , and screen the area where the pressure is not released evenly to obtain the local pressure release deviation area, where and They represent the pressure values of the i-th support point before and after turning over, N is the total number of support points, and They represent the pressure values in a specific area j, respectively, and M is the number of support points in the area; The support point optimization adjustment submodule adjusts the release order of the support points in the area based on the local pressure release deviation area, adjusts the time difference of the mattress according to the pressure gradient sorting, and recalculates the adjusted pressure mean to obtain the local pressure balanced distribution record.
7. The patient turning system based on intelligent perception according to claim 1, characterized in that: The system also includes a post-turnover support adjustment module; The support adjustment module after turning over collects the bed pressure data after turning over in real time based on the local pressure balance distribution record, calculates the pressure distribution under the current patient position, adjusts the mattress deformation and support points, and matches the new state of the patient's position.
8. The patient turning system based on intelligent perception according to claim 7, characterized in that: The post-turnover support adjustment module comprises: The pressure data acquisition submodule obtains the pressure sensing data of the support area of the bed after turning over based on the local pressure balance distribution record, records the real-time pressure values of different support points, calculates the pressure average of each area, compares the pressure changes before and after turning over, analyzes the local pressure distribution of the support points, and obtains the pressure distribution data after turning over; The support area optimization submodule calculates the pressure change rate of the differential support area based on the pressure distribution data after turning over, selects the area with uneven pressure distribution, adjusts the deformation amplitude of the corresponding support point, adjusts the support strength of each area, balances the force state of different parts, and obtains the support point optimization adjustment plan; The patient position matching submodule adjusts the support point height and deformation amplitude according to the support point optimization adjustment plan, monitors the changes in the patient's position pressure distribution in real time, calculates the balance of the optimized support point, and matches the new state of the patient's position.