Pressure damage early warning method, device and equipment for pressure sensing mattress

Abstract

The invention provides a pressure damage early warning method, device and equipment for a pressure sensing mattress, and relates to the technical field of monitoring and early warning. The method comprises the following steps: continuously acquiring and recording pressure distribution data generated by a pressure sensing mattress, wherein the pressure distribution data comprises pressure values of a plurality of discrete monitoring points on the pressure sensing mattress and a change time sequence of the pressure values; identifying a current pose of the user based on the pressure distribution data; based on the current posture of the user, determining a body part, which is easy to be pressed, of the user; for each determined body part, calculating a real-time pressed state value according to the corresponding pressure data; and comparing each real-time compression state value with a preset risk threshold value, and dynamically generating and executing an early warning instruction based on a comparison result. The time point of next early warning can be automatically adjusted, the early warning interval is shortened when the risk is high, the interval is prolonged when the risk is low, the safety of a user is guaranteed, and unnecessary nursing interference is avoided.

Description

Technical Field

[0001] This invention relates to the field of monitoring and early warning technology, and in particular to a method, device and equipment for early warning of pressure injury in a pressure-sensing mattress. Background Technology

[0002] Pressure injuries, a common complication in long-term bedridden patients, are mainly caused by impaired blood circulation due to continuous pressure on local tissues. In clinical nursing, regular turning is one of the most effective measures for preventing pressure injuries. Traditional pressure injury prevention relies entirely on manual observation and recording by nursing staff, which not only increases their workload but also often results in delayed turning due to busy schedules or shift changes, affecting the effectiveness of prevention.

[0003] In recent years, intelligent pressure-sensing mattresses have been increasingly applied in clinical practice, monitoring the user's body pressure distribution and issuing alarms when sustained pressure exceeds a preset time. However, existing methods still have significant limitations. One common approach is to set a fixed pressure threshold; when local pressure exceeds this threshold, an alarm is triggered, and if the warning lasts for a fixed duration, a full alarm is activated. Another approach categorizes static risk levels based on local pressure magnitude and specifies a corresponding turning frequency for each level. Both approaches make isolated warning judgments based on pressure and duration, failing to characterize the dynamic accumulation and dissipation of tissue pressure risk, and also failing to correlate pressure distribution with the physiological characteristics of specific anatomical locations.

[0004] Therefore, because it only identifies localized pressure values ​​rather than pressure values ​​for specific body parts, the system may still trigger a fixed alarm even after the user has turned over and adjusted, when the pressure in the previously compressed area has eased, leading to a higher false alarm rate. Conversely, if the user maintains the same posture for a long time without exceeding the pressure threshold, the system is prone to missing alarms. Furthermore, current technology lacks the ability to assess dynamic changes in pressure conditions, and the rigid warning interval settings cannot be flexibly adjusted based on the user's real-time pressure risk. This not only increases the workload of caregivers but also limits the accuracy and timeliness of warnings. Summary of the Invention

[0005] This invention provides a method, device, and equipment for early warning of pressure injury in pressure-sensing mattresses, to address the problem of low accuracy and timeliness in early warning of pressure injury risks.

[0006] In a first aspect, embodiments of the present invention provide a method for early warning of pressure injury in a pressure-sensing mattress, comprising: Continuously acquire and record pressure distribution data generated by the pressure-sensing mattress. The pressure distribution data includes the pressure values ​​and their change time sequence at multiple discrete monitoring points on the pressure-sensing mattress. Based on pressure distribution data, identify the user's current pose; Based on the user's current posture, identify the user's most vulnerable body parts; For each identified body part, a real-time pressure state value is calculated based on its corresponding pressure data; whereby the real-time pressure state value is a quantitative indicator generated based on the historical cumulative pressure effect and pressure intensity of the body part. Each real-time pressure state value is compared with a preset risk threshold, and warning commands are dynamically generated and executed based on the comparison results; the timing of the warning trigger in the warning command is adaptively adjusted according to the change of the real-time pressure state value.

[0007] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: When the number of discrete monitoring points whose pressure values ​​change more than a preset fluctuation threshold within a preset time period exceeds a preset number threshold, the user's current posture is identified based on the current pressure distribution data.

[0008] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: When the distance between the caregiver and the pressure-sensing mattress is less than a preset distance for a duration exceeding a preset duration threshold, the user's current posture is identified based on the current pressure distribution data.

[0009] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: The current pressure distribution data is matched with the pre-stored pose templates to identify the user's current pose; each pose template corresponds to the characteristics of pressure distribution data of a typical pose.

[0010] In one possible implementation, based on the user's current pose, the user's pressure-sensitive body parts are determined, including: Based on the user's current posture, the corresponding pressure-prone body parts are found in the preset posture-body part mapping relationship to determine the user's pressure-prone body parts.

[0011] In one possible implementation, for each identified body part, a real-time pressure state value is calculated based on its corresponding pressure data, including: When the pressure intensity of the first body part is greater than or equal to the first pressure threshold corresponding to the first body part, the pressure duration of the first body part is accumulated; wherein, the first body part is any determined body part. When the pressure intensity of the first body part is less than the second pressure threshold corresponding to the first body part, the decompression time of the first body part is accumulated; wherein, the second pressure threshold is less than the first pressure threshold. The real-time pressure state value of the first body part is calculated based on the duration of pressure application and the duration of decompression.

[0012] In one possible implementation, the real-time pressure state value of the first body part is calculated based on the pressure duration and decompression duration, including: If the decompression duration is greater than or equal to the preset full decompression threshold, the real-time pressure status value, pressure duration, and decompression duration of the first body part will all be reset to 0. If the decompression duration is less than the complete decompression threshold, then according to the formula... Calculate the real-time compression state value of the first body part ;in, As the first coefficient, For the duration of pressure, As the second coefficient, This refers to the duration of stress reduction.

[0013] In one possible implementation, each real-time pressure state value is compared with a preset risk threshold, and warning commands are dynamically generated and executed based on the comparison results, including: For each identified body part, the remaining time to reach the risk threshold is calculated based on the real-time pressure state value of that body part and the corresponding risk threshold; where each body part corresponds to a different risk threshold. The timing for triggering an early warning in an early warning instruction is determined based on the minimum value among the remaining durations.

[0014] In a second aspect, embodiments of the present invention provide a pressure injury early warning device for a pressure-sensing mattress, comprising: The acquisition module is used to continuously acquire and record the pressure distribution data generated by the pressure sensing mattress. The pressure distribution data includes the pressure values ​​and their change time sequence at multiple discrete monitoring points on the pressure sensing mattress. The pose recognition module is used to identify the user's current pose based on pressure distribution data; The body part determination module is used to determine the user's pressure-prone body parts based on the user's current posture. The state calculation module is used to calculate the real-time pressure state value for each determined body part based on its corresponding pressure data; the real-time pressure state value is a quantitative index generated based on the historical cumulative pressure effect and pressure intensity of the body part. The comparison and early warning module is used to compare each real-time pressure state value with a preset risk threshold, and dynamically generate and execute early warning commands based on the comparison results; the early warning triggering timing in the early warning command is adaptively adjusted according to the change of the real-time pressure state value.

[0015] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0016] The pressure injury early warning method, device, and equipment for pressure-sensing mattresses provided in this invention continuously monitor pressure distribution data, identify the user's current posture and determine vulnerable body parts, and then calculate the real-time pressure state value of the body parts based on pressure values ​​and duration at multiple locations on the pressure-sensing mattress. This comprehensively considers the historical cumulative pressure effect and current pressure intensity, accurately quantifying the urgency and rate of risk accumulation for each body part to reach a critical risk state, thereby dynamically updating the early warning interval and significantly improving the accuracy and personalization of the early warning. Instead of relying on fixed pressure thresholds or timers, it can automatically adjust the timing of the next warning, shortening the warning interval when the risk is high and extending it when the risk is low, ensuring user safety and avoiding unnecessary nursing interference. Finally, this method reduces the burden on nursing staff, helping them improve work efficiency and quality of care through intuitive pressure state values ​​and scientific early warning intervals. Attached Figure Description

[0017] Figure 1 This is an application scenario diagram of the pressure injury early warning method for pressure-sensing mattresses provided in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the implementation of a pressure injury early warning method for a pressure-sensing mattress provided in an embodiment of the present invention. Figure 3 This is a structural diagram of a pressure injury early warning device for a pressure-sensing mattress provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] Figure 1 This is an application scenario diagram of the pressure injury early warning method for pressure-sensing mattresses provided in the embodiments of the present invention, such as... Figure 1As shown, this method can be applied to situations requiring pressure injury prevention for bedridden users, such as clinical inpatient wards, long-term care facilities, or home-based care. Once the user lies on a bed equipped with a pressure-sensing mattress, the system begins unattended continuous operation, continuously collecting real-time pressure distribution data. Combined with built-in programs, it automatically identifies the user's posture, locates pressure-prone areas, calculates real-time pressure status values, and performs dynamic risk assessment. Finally, an adaptive early warning mechanism pushes warning information to nursing staff, supporting timely interventions such as turning the user over.

[0020] See Figure 2 The diagram illustrates the implementation flowchart of the pressure injury early warning method for a pressure-sensing mattress provided by an embodiment of the present invention, which is described in detail below: Step 201: Continuously acquire and record the pressure distribution data generated by the pressure-sensing mattress. The pressure distribution data includes the pressure values ​​and their change sequence at multiple discrete monitoring points on the pressure-sensing mattress.

[0021] In this embodiment, this step is achieved through a pressure sensor array uniformly or non-uniformly distributed on the mattress surface and a corresponding data acquisition and storage module. Pressure readings from each sensing unit are continuously collected at a certain sampling frequency, thereby obtaining pressure distribution data that reflects the magnitude and temporal sequence of pressure changes in different areas of the user's body. The sampling frequency can be set according to the general requirements of pressure change resolution in the clinical field, for example, it can be set to once per second. In the following description, the entire assembly of the pressure sensor array, corresponding data acquisition and storage module, and processor on the pressure-sensing mattress is referred to as the "system".

[0022] By continuously acquiring pressure distribution data at different times, a dynamic, spatiotemporally correlated data field can be formed, providing a comprehensive and objective source of physical information for subsequent analysis. Compared to simple single-point threshold judgment, this data field can support the capture of potential risks from overall distribution characteristics and time-series change trends.

[0023] Step 202: Identify the user's current pose based on pressure distribution data.

[0024] In this embodiment, different body postures (such as supine, lateral, and prone) result in distinct distribution patterns of body weight on the mattress, specifically manifested in differences in pressure center coordinates, overall shape of pressure projection, and the number and location distribution of high-pressure zones. Based on this, the user's current body posture can be inferred by analyzing the spatial and statistical characteristics of the pressure distribution data.

[0025] Specifically, a set of predefined feature parameters can be extracted from the current pressure distribution data, such as the centroid coordinates of the pressure matrix, moment of inertia, and the number and area of ​​connected high-pressure regions. These feature parameters are then input into a classifier for recognition, or matched with a preset template for similarity. The result with the highest matching degree is selected as the user's current pose. Specifically, the classifier can be a support vector machine (SVM), which needs to be trained before using it to recognize poses. During training, in a controlled environment, multiple test subjects are arranged to maintain predefined standard body positions (supine, left lateral, right lateral, prone, etc.) in sequence. Test subjects can include different body types, and stable pressure distribution data is collected using a target pressure-sensing mattress. Predefined feature parameters are extracted from each pressure distribution data point to form a data sample, and the corresponding standard body position is used as a label to form the training dataset. The types and number of feature parameters can be adjusted according to the actual results. To improve the efficiency and effectiveness of subsequent training, the data samples can also be standardized to unify the numerical scale of each feature.

[0026] The training dataset is then input into the support vector machine algorithm for training. The radial basis function is used as the kernel function, and the penalty coefficient C is set to a value range of 1-10. The gamma value of the radial basis function is set to a value range of 0.01-0.1. The internal parameters of the support vector machine algorithm are optimized through cross-validation until the recognition accuracy reaches the required level. The training is then completed, and the trained support vector machine can be used for pose recognition.

[0027] By using pressure distribution data for posture recognition, physical pressure signals can be automatically and in real time converted into postural information with clear nursing significance, providing key context for subsequent personalized analysis of specific postures.

[0028] Step 203: Based on the user's current posture, determine the user's vulnerable body parts.

[0029] In this embodiment, based on the identified current posture and combined with knowledge of human anatomy, the anatomical sites most likely to bear greater pressure and thus have a higher risk of pressure injury under that specific posture can be deduced. The principle is that the main weight-bearing bony prominences and contact surfaces of the human body change systematically under different postures. For example, when the identification result is "supine position," the system can deduce that the sacrum and heel are the main risk points; when the identification result is "left lateral decubitus position," it can deduce that the left shoulder and left hip are the main risk points.

[0030] In practical implementation, the system can have a built-in logical mapping table, whose input is pose identifiers and output is a set of one or more body part identifiers. The preset logical mapping table is the knowledge base for this reasoning, and its function is to transform general nursing knowledge into a judgment basis that the system can execute. It can be obtained in advance by summarizing and generalizing authoritative guidelines for the prevention of pressure injuries and biomechanical consensus.

[0031] The beneficial effect of this step is that it shifts the system's monitoring focus from indiscriminate high-pressure areas to target sites with physiological and clinical significance, achieving precise location of risk assessment and providing a clear object for subsequent detailed calculations.

[0032] Step 204: For each determined body part, calculate the real-time pressure state value based on its corresponding pressure data; wherein, the real-time pressure state value is a quantitative index generated based on the historical cumulative pressure effect and pressure intensity of the body part.

[0033] In this embodiment, the real-time pressure status value serves as a quantitative and dynamic risk indicator. Its value directly represents the urgency of the tissue approaching the ischemic injury threshold. This indicator not only reflects the instantaneous pressure situation but also encompasses the pressure history of the area and the risk mitigation effect brought about by turning over to relieve pressure, thus more scientifically representing the real-time urgency of tissue damage risk.

[0034] The core principle of calculating real-time pressure state values ​​is to establish a dynamic process simulation of risk accumulation and dissipation. Specifically, for each body part, its estimated projection area in the current pose can be determined according to a preset logical mapping table. Then, indicators characterizing the pressure load of this area (such as the average pressure, peak pressure, or pressure-area integral value of the sensors in this area) are extracted from the pressure distribution data as the current pressure intensity of that body part. The body part can be a predefined anatomical region prone to pressure injury, such as the sacral region, ischial tuberosity region, or heel region. The system then combines the current pressure intensity with the historical pressure sequence of that part (i.e., information characterizing the historical cumulative effect of pressure) and calculates a new real-time pressure state value through a state update function. This function can adopt a decaying memory approach, weighting and fusing historical effects with the current intensity. For example, the real-time pressure state value can be the current pressure intensity plus a time-domain weighted integral result based on historical pressure data.

[0035] Taking a simplified implementation as an example, a decaying memory approach is used for state updates. Specifically, an accumulator is maintained for each monitored area. When the average pressure in that area exceeds a certain empirical threshold, the accumulator increments over time; when the pressure falls below another threshold, the accumulator decrements at a different rate. This setting considers the increase and decrease in tissue ischemia under different pressures in different body parts. The increment / decrement rate, threshold, and other parameters are key preset data. These parameters can be set to default values ​​based on generally accepted evidence-based medicine conclusions, and an interface for adjustment via the "Personalized Settings" button is provided. Nursing staff can click the "High Risk," "Medium Risk," or "Low Risk" button based on the user's Braden score or visual skin condition. The system will automatically load the corresponding parameter combination, thereby achieving differentiated calculations for users with different risk levels.

[0036] Step 205: Compare each real-time pressure state value with a preset risk threshold, and dynamically generate and execute early warning instructions based on the comparison results; wherein, the early warning triggering timing in the early warning instructions is adaptively adjusted according to the changes in the real-time pressure state value.

[0037] In this embodiment, the system's early warning decision is based on a dynamic assessment of real-time pressure status values. Specifically, the system compares the real-time pressure status value of each body part with one or more preset risk thresholds for that body part.

[0038] Furthermore, the system can analyze the trend of real-time pressure status value over time, such as calculating its rate of increase. Combining the current real-time pressure status value with the trend of change, it can predict the time required for the real-time pressure status value to reach the risk threshold and set the predicted time as the timing for the next early warning check or early warning trigger.

[0039] Risk thresholds and algorithms used for trend prediction are the core of dynamic early warning mechanisms, enabling a leap from static threshold alarms to intelligent alarms based on risk prediction. The risk threshold can be set based on clinical risk levels; the trend algorithm can be selected using signal processing techniques, such as linear extrapolation or filtered prediction.

[0040] This early warning system enables intelligent and adaptive triggering of warnings: when risks accumulate rapidly, warnings become more frequent and urgent; when risks are stable or declining, the warning interval (the time difference between the current moment and the warning trigger point) automatically extends. This minimizes unnecessary nursing disruptions, optimizes the allocation of nursing resources, and improves the user experience while ensuring user safety.

[0041] The updated warning interval will be loaded into a countdown timer. When the countdown reaches zero, the system will trigger a warning action. Warning methods may include flashing bedside indicator lights, sounding a buzzer, and sending a notification message to the nurses' station. Furthermore, the system supports selecting different warning intensities via a "Warning Mode" button. For example, "Silent Mode" triggers only a light alert, "Standard Mode" uses a combination of sound and light alerts, and "Enhanced Mode" adds a warning message to the nurse's mobile terminal in addition to the sound and light alerts.

[0042] This invention continuously monitors pressure distribution data, using it to identify the user's current posture and determine pressure-prone body parts. Then, based on pressure values ​​and duration at multiple locations on the pressure-sensing mattress, it calculates the real-time pressure status of each body part. This comprehensively considers historical pressure accumulation benefits and current pressure intensity, accurately quantifying the urgency and rate of risk accumulation for each part to reach a critical risk state. This dynamically updates the warning interval, significantly improving the accuracy and personalization of warnings. Instead of relying on fixed pressure thresholds or timers, it automatically adjusts the timing of the next warning, shortening the warning interval when the risk is high and extending it when the risk is low, ensuring user safety and avoiding unnecessary nursing interventions. Finally, this method reduces the burden on nursing staff, helping them improve work efficiency and the quality of care through intuitive pressure status values ​​and scientific warning intervals.

[0043] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: When the number of discrete monitoring points whose pressure values ​​change more than a preset fluctuation threshold within a preset time period exceeds a preset number threshold, the user's current posture is identified based on the current pressure distribution data.

[0044] In this embodiment, a change detection mechanism for triggering pose re-identification is provided. Its principle is to monitor the overall stability of the pressure distribution, using large-scale, significant pressure change events as indicators that pose may have changed.

[0045] Specifically, the system continuously monitors pressure distribution data and calculates the pressure changes at each monitoring point between adjacent time points. First, it counts the number of monitoring points whose pressure changes exceed a preset fluctuation threshold within a continuous preset time period. If this number consistently exceeds the preset threshold, it indicates that a systemic movement or turning over may have occurred, meaning the user's posture may have changed. For example, if the preset time period is 5 seconds, the preset fluctuation threshold is 8 mmHg, and the preset threshold is 15% of the total number of monitoring points, if the number of monitoring points with pressure changes exceeding 8 mmHg reaches 25% of the total number of monitoring points within a 5-second time window, it is determined that a systemic movement or turning over may have occurred.

[0046] Once a potential change in posture is detected, the posture recognition function is immediately invoked to perform a new posture recognition based on the latest and stable pressure distribution data. By setting preset time periods, preset fluctuation thresholds, and preset quantity thresholds, it is possible to effectively distinguish between localized micro-movements and systemic postural changes. The preset time periods, preset fluctuation thresholds, and preset quantity thresholds can be determined based on statistical analysis of pressure signals from common body movements and turning actions.

[0047] This approach provides the system with an event trigger condition, eliminating the need for continuous high-consumption pose recognition calculations. The recognition process is only initiated when an important event that may change the body position context is detected, which can significantly optimize the allocation of computing resources and real-time performance of the system.

[0048] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: When the distance between the caregiver and the pressure-sensing mattress is less than a preset distance for a duration exceeding a preset duration threshold, the user's current posture is identified based on the current pressure distribution data.

[0049] In this embodiment, the determination of pose changes can also be aided by incorporating external environmental context information. The close approach of caregivers is used as a strong correlation signal that nursing procedures such as assisted turning are about to occur or are in progress.

[0050] Specifically, the system uses an additional personnel positioning module (such as infrared, UWB, or Bluetooth beacons) to detect whether a caregiver has entered a range centered on the mattress with a preset radius (e.g., 1.5 meters). When a caregiver is detected entering this range and remains there for more than a preset duration threshold (e.g., 10 seconds), the system infers a high probability that a positional adjustment activity led by the caregiver has occurred, or that other non-turning care procedures have been performed. Since assisted turning can sometimes be smooth, and non-turning care procedures can also cause changes in the pressure intensity on the user's body parts, the system triggers a pose recognition process based on the latest pressure distribution data, regardless of whether the pressure data shows a drastic change.

[0051] The preset distance and preset duration thresholds together define the detection conditions for effective nursing operations, which can reduce misjudgments caused by nursing staff simply passing by the bedside and improve the accuracy of detecting planned nursing turning events. The values ​​are determined based on observation and time measurement of typical nursing workflows.

[0052] This embodiment adds conditions for judging possible changes in posture, enabling the system to more intelligently integrate with actual clinical scenarios. Even under conditions of mild stress changes, it can update posture information in a timely manner, ensuring that risk assessment is always based on the correct postural context. Since continuous posture recognition would lead to significant system overhead, this embodiment reduces system overhead by setting trigger conditions to recognize the user's posture when these conditions are met. Furthermore, it can recognize posture when the user may be turning over, ensuring that the accuracy of the warning is not affected.

[0053] In one possible implementation, the user's current pose is identified based on pressure distribution data, including: The current pressure distribution data is matched with the pre-stored pose templates to identify the user's current pose; each pose template corresponds to the characteristics of pressure distribution data of a typical pose.

[0054] In this embodiment, a large number of typical pressure distribution patterns under standard body positions (supine, left lateral, right lateral, etc.) can be pre-collected in a controlled environment to establish a database of posture templates. Then, the real-time pressure distribution data is compared with a series of predefined reference patterns (i.e., posture templates) representing standard body positions for similarity.

[0055] Each pose template can be a typical pressure distribution matrix or a set of feature vectors extracted from a typical distribution. During recognition, the system calculates a matching score between the current pressure distribution data and each pre-stored pose template, for example, by calculating cross-correlation, mean squared error, or more advanced image / pattern similarity metrics. The system selects the pose template with the highest matching score and outputs the body position it represents as the recognition result.

[0056] This embodiment provides an intuitive, interpretable, and computationally feasible pose recognition method that can map complex pressure distributions to discrete, meaningful body position categories.

[0057] In one possible implementation, based on the user's current pose, the user's pressure-sensitive body parts are determined, including: Based on the user's current posture, the corresponding pressure-prone body parts are found in the preset posture-body part mapping relationship to determine the user's pressure-prone body parts.

[0058] In this embodiment, the posture-body location mapping relationship is a pre-defined knowledge base that clearly defines which body parts are at high risk in each standard body position, such as the sacrum and coccyx and heels corresponding to the supine position. This allows the user's high-risk pressure locations to be identified from the current pressure distribution data, and the pressure intensity of those locations can be extracted from the current pressure distribution data.

[0059] This mapping relationship can be a lookup table, a configuration file, or a set of records in a database. Once the system obtains the current pose, it uses that pose as a keyword or index to query within the pre-defined mapping relationship. The query results directly return one or more body part identifiers (such as "sacrum," "left ischial tuberosity," or "right heel"). These identifiers represent the pressure-prone body parts that require close monitoring in the current pose.

[0060] The core of the position-site mapping relationship is to associate a user's specific body posture (such as supine, left lateral, right lateral, and prone) with the main weight-bearing and pressure-prone anatomical sites (such as the sacrum, heel, greater trochanter, and acromion) in the current posture, and further describe the typical spatial distribution patterns and relative pressure intensity characteristics of these sites on the pressure sensing array. Specifically, in the supine position, the main pressure-bearing sites associated with the mapping relationship are the occipital bone, scapular region, sacral region, and heel. The pressure distribution usually presents a multi-peak pattern along the body's central axis, with the sacral region often having the core peak pressure and the heel being a significant pressure concentration point. In the lateral position (taking the right lateral position as an example), the mapping relationship is associated with the right auricle, acromion, greater trochanter, lateral knee joint, and ankle. The pressure characteristics are manifested as a longitudinal pressure band along one side of the body, with the greater trochanter region often having the absolute peak pressure and the acromion region having a stable secondary peak. In the prone position, the mapping relationship involves the forehead, lower sternum, anterior superior iliac spine, knee, and dorsum of the foot. The center of pressure shifts forward, and the lower sternum and anterior superior iliac spine region usually bear the main load.

[0061] The accuracy of this mapping relationship is the foundation of dynamic risk assessment; therefore, the system design can incorporate a mechanism for personalized adjustment. During initialization or care plan development, caregivers can calibrate and select preset mapping parameters (such as the center position and size of the pressure zone) based on the user's actual body type (such as height, weight, and build) and the characteristics of the mattress used (such as firmness).

[0062] Furthermore, to extract the pressure intensity of each body part, the system needs to predefine several body parts that need to be monitored and define a fixed logical region corresponding to each part on the pressure sensor array. For example, this can be preset based on mattress size and standard ergonomic positions. During calculation, the system directly extracts the real-time pressure information of these preset regions from the pressure distribution data as the pressure intensity of each part.

[0063] In another approach, the pose-body mapping relationship, besides specifying the pressure-bearing parts in a particular posture, can also embed the relative pressure intensity relationship between the main weight-bearing parts in different postures. For example, this relationship can clearly state that "when lying on one's side, the pressure in the shoulder and hip joint areas is significantly higher than in other parts, and is often the extreme point in the pressure distribution." Based on this knowledge, after the system identifies that the user is currently in a lying position, it can actively search for pressure peak points in the global pressure distribution data. The extraction logic is as follows: first, scan the entire pressure matrix, identify the local pressure maximum point and its value, and then, according to a preset relationship (such as "the maximum pressure point in the lying position corresponds to the shoulder"), directly identify the peak value as the pressure intensity of the shoulder.

[0064] If the mapping relationship further defines multiple distinct high-pressure areas in this posture (e.g., the shoulder first, the hip second), the system can sequentially extract the first and second largest effective pressure peaks globally and assign them to the shoulder and hip respectively. To ensure accuracy, this process typically combines simple spatial constraints (e.g., the peak point must be located in the upper half of the mattress to be considered the shoulder) and numerical threshold filtering (e.g., the peak value must be significantly higher than the background pressure average) to eliminate interference caused by sensor noise or minor movements. Ultimately, these extracted peak pressure values ​​are used as the pressure intensity input to the subsequent time accumulation and state value calculation modules for the corresponding anatomical locations. This method does not require pre-setting fixed sensor areas for each location but dynamically matches them based on real-time pressure distribution patterns and pose knowledge, enhancing the system's adaptability to different body types and lying postures.

[0065] In one possible implementation, for each identified body part, a real-time pressure state value is calculated based on its corresponding pressure data, including: When the pressure intensity of the first body part is greater than or equal to the first pressure threshold corresponding to the first body part, the pressure duration of the first body part is accumulated; wherein, the first body part is any determined body part. When the pressure intensity of the first body part is less than the second pressure threshold corresponding to the first body part, the decompression time of the first body part is accumulated; wherein, the second pressure threshold is less than the first pressure threshold. The real-time pressure state value of the first body part is calculated based on the duration of pressure application and the duration of decompression.

[0066] In this embodiment, two pressure thresholds are introduced to control the timing of the pressure and decompression states respectively, and the accumulation of risk is characterized by the calculation of the duration. The principle is to define two pressure level limits for each body part: the first pressure threshold is a higher threshold used to determine whether it is in a state of significant risk accumulation; the second pressure threshold is a lower threshold used to determine whether it is in a state of effective risk mitigation.

[0067] The system continuously monitors the pressure intensity calculated from the corresponding area of ​​the location. When the pressure intensity is greater than or equal to the first pressure threshold, the system considers the location to be under a clear high-pressure risk and thus begins or continues to accumulate the pressure duration. When the pressure intensity is less than the second pressure threshold, the system considers the location to be in an effective low-pressure recovery period and thus begins or continues to accumulate the decompression duration.

[0068] The calculation of the real-time pressure state value is based on these two durations, which can be, for example, a linear or nonlinear combination of them. The first and second pressure thresholds are key parameters in this embodiment, defining the critical conditions for risk accumulation and dissipation. They are obtained based on physiological parameters such as tissue capillary closure pressure and clinical experience, and can be configured differently for different locations. This method, through dual-threshold hysteresis comparison, avoids frequent timer switching caused by pressure value fluctuations around a single threshold, making the accumulation of pressure and decompression durations more stable. This results in a more smooth and accurate reflection of the net accumulation trend of risk in the calculated state value.

[0069] In one possible implementation, the real-time pressure state value of the first body part is calculated based on the pressure duration and decompression duration, including: If the decompression duration is greater than or equal to the preset full decompression threshold, the real-time pressure status value, pressure duration, and decompression duration of the first body part will all be reset to 0. If the decompression duration is less than the complete decompression threshold, then according to the formula... Calculate the real-time compression state value of the first body part ;in, As the first coefficient, For the duration of pressure, As the second coefficient, This refers to the duration of stress reduction.

[0070] In this embodiment, the system presets a complete decompression threshold for each monitored site, representing the minimum time required for that site to remain in a low-pressure state (pressure below a second pressure threshold) to achieve complete recovery. The principle is to simulate the physiological process by which previously accumulated ischemic damage risk can completely dissipate after sufficient rest, thus introducing a mechanism for complete risk reset.

[0071] Before each calculation, first check the cumulative decompression time for that area. .if If the value is greater than or equal to the complete decompression threshold, the area is considered to have fully recovered from the previous pressure, and its risk status is reset to zero, i.e., the status value S and the pressure duration are reset. and stress reduction duration Reset everything to 0 to clear historical risks. If If the value is below this threshold, the recovery is considered insufficient, and historical risks still have residual effects. In this case, the formula is applied. The calculation is performed, where a and b are the first and second coefficients, respectively. The multiplication operation in the formula reflects the linear superposition and offsetting relationship between the impact of duration on risk.

[0072] The complete decompression threshold and the first coefficient 'a' and second coefficient 'b' required for calculating the state value are the core parameters of this embodiment. The complete decompression threshold can be derived based on animal experiments, clinical observation data fitting, or theoretical models. Coefficients 'a' and 'b' represent the rate of risk accumulation under pressure and the rate of recovery under decompression, respectively, and can be set to default values ​​based on evidence-based medicine data at the system's factory, for example, a=1 and b=0.8. They are designed to be quickly adjusted via the "Risk Level" button on the nurse's interface. For example, if a nurse assesses a user as high-risk (low Braden score), pressing the "High Risk" button will automatically increase the value of 'a' and decrease the value of 'b', for example, a=1.2 and b=0.6, making the model more sensitive to changes in the user's risk.

[0073] The beneficial effects of this embodiment are that it makes the risk assessment model more in line with physiological laws, avoids the infinite accumulation of historical risks, and provides a clear and quantifiable mathematical expression, which is easy to implement and adjust.

[0074] In one possible implementation, each real-time pressure state value is compared with a preset risk threshold, and warning commands are dynamically generated and executed based on the comparison results, including: For each identified body part, the remaining time to reach the risk threshold is calculated based on the real-time pressure state value of that body part and the corresponding risk threshold; where each body part corresponds to a different risk threshold. The timing for triggering an early warning in an early warning instruction is determined based on the minimum value among the remaining durations.

[0075] In this embodiment, a unique state threshold needs to be preset for each part. After calculating the real-time pressure state value S of a certain part, the system immediately calculates the remaining time for that part. A direct calculation method is: Remaining time = (State threshold of the part - S) / R. Here, R is a risk approach rate parameter, which can be a fixed constant, such as the coefficient a in the previous embodiment; or it can be a value dynamically estimated based on the recent state value change trend of the part. The physical meaning of this calculation result is: if the current risk change trend continues, it is expected that after the remaining time, the state value S will reach the warning critical point (risk threshold). The calculated remaining time is the warning countdown updated for that part.

[0076] Furthermore, setting differentiated complete decompression thresholds and status thresholds for different body parts is a core step in achieving precise and personalized pressure injury early warning. Specifically, the system first incorporates a default threshold template based on clinical consensus, which is directly linked to the anatomical vulnerability and frequency of pressure injuries in different body parts. For example, for the heel and sacrum, where subcutaneous tissue is thin and blood supply is poor, the system assigns a shorter complete decompression threshold (e.g., 15-20 minutes) and a lower risk threshold (i.e., a smaller allowable cumulative risk) to implement stricter protection. For the scapular region or back, where muscle and soft tissue are relatively abundant, the default thresholds can be relatively more relaxed. When a specific user is introduced, the system automatically scales the default thresholds for all body parts globally based on the overall risk level entered by the caregiver (e.g., "high risk" selected based on the Braden score). For example, all thresholds for a "high risk" user may be adjusted to 80% of the default value, thus achieving rapid matching of risk level and monitoring intensity.

[0077] In addition, the system can be configured with a settings button to receive configuration operations from nursing staff regarding warning-related information. To improve the accuracy of subsequent analysis, during system initialization or when used for the first time by a user, nursing staff can be guided to input the user's basic body type classification, such as "standard," "underweight," or "obese," through the settings button on the user interface. Based on this classification, additional distribution weighting or baseline compensation can be applied to the pressure readings of underweight or obese users to more accurately reproduce their true body pressure distribution, avoid sensing bias caused by body type differences, and thus make the monitoring data more representative of individuals.

[0078] The system allows for targeted fine-tuning for special circumstances. Caregivers can manually adjust thresholds for specific areas via the user management interface. For example, if a user has developed a stage I pressure injury in the left greater trochanter region, the complete decompression threshold for that area can be significantly shortened, and the status threshold lowered to enhance monitoring of that point. Conversely, for areas that cannot withstand pressure for any reason (such as the ankle with an external fixator), the first pressure threshold for that area can be set to an extremely low value, making almost any minute pressure from the system considered a risk accumulation.

[0079] For situations involving multiple pressure-prone body parts, the system runs multiple monitoring processes in parallel, each corresponding to one body part and independently calculating the remaining time for that part. The system maintains a global pressure injury warning interval. The update strategy for this interval is to continuously scan the calculated remaining time values ​​for all parts and directly set the minimum value among these values ​​as the new global warning interval.

[0080] For example, if the remaining time for the sacrococcygeal region is 70 minutes, for the left heel it's 90 minutes, and for the right heel it's 45 minutes, the system will update the warning interval to 45 minutes. This strategy of taking the minimum value is a conservative and safe design principle, ensuring that the system warning will be triggered if the risk in any single area reaches the critical point ahead of time.

[0081] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0082] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0083] Figure 3 A schematic diagram of a pressure injury early warning device for a pressure-sensing mattress provided in an embodiment of the present invention is shown. For ease of explanation, only the parts relevant to the embodiment of the present invention are shown, and are described in detail below: like Figure 3 As shown, the pressure injury early warning device 3 for a pressure-sensing mattress includes: The acquisition module 31 is used to continuously acquire and record the pressure distribution data generated by the pressure sensing mattress. The pressure distribution data includes the pressure values ​​and their change sequence at multiple discrete monitoring points on the pressure sensing mattress. The pose recognition module 32 is used to identify the user's current pose based on pressure distribution data; The body part determination module 33 is used to determine the user's pressure-prone body parts based on the user's current posture. The state calculation module 34 is used to calculate the real-time pressure state value for each determined body part based on its corresponding pressure data; wherein, the real-time pressure state value is a quantitative index generated based on the historical cumulative pressure effect and pressure intensity of the body part. The comparison and early warning module 35 is used to compare each real-time pressure state value with a preset risk threshold, and dynamically generate and execute early warning instructions based on the comparison results; wherein, the early warning triggering timing in the early warning instruction is adaptively adjusted according to the change of the real-time pressure state value.

[0084] In one possible implementation, the pose recognition module 32 is specifically used for: When the number of discrete monitoring points whose pressure values ​​change more than a preset fluctuation threshold within a preset time period exceeds a preset number threshold, the user's current posture is identified based on the current pressure distribution data.

[0085] In one possible implementation, the pose recognition module 32 is specifically used for: When the distance between the caregiver and the pressure-sensing mattress is less than a preset distance for a duration exceeding a preset duration threshold, the user's current posture is identified based on the current pressure distribution data.

[0086] In one possible implementation, the pose recognition module 32 is specifically used for: The current pressure distribution data is matched with the pre-stored pose templates to identify the user's current pose; each pose template corresponds to the characteristics of pressure distribution data of a typical pose.

[0087] In one possible implementation, the location determination module 33 is specifically used for: Based on the user's current posture, the corresponding pressure-prone body parts are found in the preset posture-body part mapping relationship to determine the user's pressure-prone body parts.

[0088] In one possible implementation, the state calculation module 34 is specifically used for: When the pressure intensity of the first body part is greater than or equal to the first pressure threshold corresponding to the first body part, the pressure duration of the first body part is accumulated; wherein, the first body part is any determined body part. When the pressure intensity of the first body part is less than the second pressure threshold corresponding to the first body part, the decompression time of the first body part is accumulated; wherein, the second pressure threshold is less than the first pressure threshold. The real-time pressure state value of the first body part is calculated based on the duration of pressure application and the duration of decompression.

[0089] In one possible implementation, the state calculation module 34 is specifically used for: If the decompression duration is greater than or equal to the preset full decompression threshold, the real-time pressure status value, pressure duration, and decompression duration of the first body part will all be reset to 0. If the decompression duration is less than the complete decompression threshold, then according to the formula... Calculate the real-time compression state value of the first body part ;in, As the first coefficient, For the duration of pressure, As the second coefficient, This refers to the duration of stress reduction.

[0090] In one possible implementation, the comparison and early warning module 35 is specifically used for: For each identified body part, the remaining time to reach the risk threshold is calculated based on the real-time pressure state value of that body part and the corresponding risk threshold; where each body part corresponds to a different risk threshold. The timing for triggering an early warning in an early warning instruction is determined based on the minimum value among the remaining durations.

[0091] This invention continuously monitors pressure distribution data, using it to identify the user's current posture and determine pressure-prone body parts. Then, based on pressure values ​​and duration at multiple locations on the pressure-sensing mattress, it calculates the real-time pressure status of each body part. This comprehensively considers historical pressure accumulation benefits and current pressure intensity, accurately quantifying the urgency and rate of risk accumulation for each part to reach a critical risk state. This dynamically updates the warning interval, significantly improving the accuracy and personalization of warnings. Instead of relying on fixed pressure thresholds or timers, it automatically adjusts the timing of the next warning, shortening the warning interval when the risk is high and extending it when the risk is low, ensuring user safety and avoiding unnecessary nursing interventions. Finally, this method reduces the burden on nursing staff, helping them improve work efficiency and the quality of care through intuitive pressure status values ​​and scientific warning intervals.

[0092] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. When the processor 40 executes the computer program 42, it implements the steps in the various method embodiments described above. Alternatively, when the processor 40 executes the computer program 42, it implements the functions of each module / unit in the various device embodiments described above.

[0093] For example, computer program 42 may be divided into one or more modules / units, which are stored in memory 41 and executed by processor 40 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 42 in electronic device 4.

[0094] Electronic device 4 may include, but is not limited to, processor 40 and memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0095] The processor 40 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0096] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard disk or RAM. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 41 can include both internal and external storage units of the electronic device 4. The memory 41 is used to store the computer program 42 and other programs and data required by the device 4. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0097] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0098] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0099] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0100] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0101] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0102] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for early warning of pressure injury in a pressure-sensing mattress, characterized in that, include: Continuously acquire and record pressure distribution data generated by the pressure-sensing mattress, the pressure distribution data including pressure values ​​and their change sequence at multiple discrete monitoring points on the pressure-sensing mattress; Based on the pressure distribution data, the user's current pose is identified; Based on the user's current posture, determine the user's vulnerable body parts; For each identified body part, a real-time pressure state value is calculated based on its corresponding pressure data; wherein, the real-time pressure state value is a quantitative index generated based on the historical cumulative pressure effect and pressure intensity of the body part. Each real-time pressure state value is compared with a preset risk threshold, and an early warning command is dynamically generated and executed based on the comparison result; wherein, the early warning triggering timing in the early warning command is adaptively adjusted according to the change of the real-time pressure state value.

2. The method for early warning of pressure injury in a pressure-sensing mattress according to claim 1, characterized in that, The step of identifying the user's current pose based on the pressure distribution data includes: When the number of discrete monitoring points whose pressure values ​​change more than a preset fluctuation threshold within a preset time period exceeds a preset number threshold, the user's current posture is identified based on the current pressure distribution data.

3. The method for early warning of pressure injury in a pressure-sensing mattress according to claim 1, characterized in that, The step of identifying the user's current pose based on the pressure distribution data includes: When the distance between the caregiver and the pressure-sensing mattress is less than a preset distance for a duration longer than a preset duration threshold, the user's current posture is identified based on the current pressure distribution data.

4. The method for early warning of pressure injury in a pressure-sensing mattress according to claim 1, characterized in that, The step of identifying the user's current pose based on the pressure distribution data includes: The current pressure distribution data is matched with the pre-stored pose templates to identify the user's current pose; each pose template corresponds to the characteristics of pressure distribution data of a typical pose.

5. The method for early warning of pressure injury in a pressure-sensing mattress according to claim 1, characterized in that, The step of determining the user's pressure-prone body parts based on the user's current posture includes: Based on the user's current posture, the corresponding pressure-prone body parts are found in the preset posture-body part mapping relationship to determine the user's pressure-prone body parts.

6. The method for early warning of pressure injury in a pressure-sensing mattress according to any one of claims 1 to 5, characterized in that, The calculation of the real-time pressure state value for each identified body part based on its corresponding pressure data includes: When the pressure intensity of the first body part is greater than or equal to the first pressure threshold corresponding to the first body part, the pressure duration of the first body part is accumulated; wherein, the first body part is any determined body part; When the pressure intensity on the first body part is less than the second pressure threshold corresponding to the first body part, the decompression time of the first body part is accumulated; wherein, the second pressure threshold is less than the first pressure threshold; Based on the duration of pressure application and the duration of decompression, the real-time pressure state value of the first body part is calculated.

7. The method for early warning of pressure injury in a pressure-sensing mattress according to claim 6, characterized in that, The calculation of the real-time pressure state value of the first body part based on the pressure duration and the decompression duration includes: If the decompression duration is greater than or equal to the preset full decompression threshold, then the real-time pressure state value of the first body part, the pressure duration, and the decompression duration are all reset to 0. If the decompression duration is less than the complete decompression threshold, then according to the formula... Calculate the real-time pressure state value of the first body part. ;in, As the first coefficient, The duration of pressure application, As the second coefficient, The decompression duration is [the specified duration].

8. The method for early warning of pressure injury in a pressure-sensing mattress according to any one of claims 1 to 5, characterized in that, The step of comparing each real-time pressure state value with a preset risk threshold, and dynamically generating and executing early warning commands based on the comparison results, includes: For each identified body part, the remaining time to reach the risk threshold is calculated based on the real-time pressure state value of that body part and the corresponding risk threshold; where each body part corresponds to a different risk threshold. The timing for triggering an early warning in an early warning instruction is determined based on the minimum value among the remaining durations.

9. A pressure injury early warning device for a pressure-sensing mattress, characterized in that, include: The acquisition module is used to continuously acquire and record the pressure distribution data generated by the pressure sensing mattress. The pressure distribution data includes the pressure values ​​and their change sequence at multiple discrete monitoring points on the pressure sensing mattress. The pose recognition module is used to identify the user's current pose based on the pressure distribution data; The body part determination module is used to determine the user's pressure-prone body parts based on the user's current posture. The state calculation module is used to calculate the real-time pressure state value for each determined body part based on its corresponding pressure data; wherein, the real-time pressure state value is a quantitative index generated based on the historical cumulative pressure effect and pressure intensity of the body part. The comparison and early warning module is used to compare each real-time pressure state value with a preset risk threshold, and dynamically generate and execute early warning instructions based on the comparison results; wherein, the early warning triggering timing in the early warning instruction is adaptively adjusted according to the change of the real-time pressure state value.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images