A complication risk early warning and nursing intervention system for patients with nephrotic syndrome

Abstract

The present application relates to the technical field of medical Internet of Things, and in particular to a complication risk early warning and nursing intervention system for nephrotic syndrome patients. The present application obtains a blood concentration risk multiplier of a current monitoring period based on a serum albumin detection value and historical liquid delivery data of a target patient. A local high-pressure point coordinate set of each monitoring period is constructed based on a pressure array matrix collected by an intelligent mattress. According to the coordinate difference between the local high-pressure point coordinate set of the current monitoring period and that of the historical monitoring period of the last effective turning over, an optimal matching point and a pairing distance are obtained, from which a local high-pressure point change ratio is obtained to analyze whether the target patient has an effective turning over. When no effective turning over occurs, the local high-pressure point change ratio and the blood concentration risk multiplier are used to analyze the pressure risk degree of the current monitoring period, and in combination with the pressure risk accumulation of the adjacent last monitoring period of the current monitoring period, a risk early warning result and a nursing intervention instruction are determined.

Description

Technical Field

[0001] This invention relates to the field of medical Internet of Things (IoT) technology, specifically to a complication risk warning and nursing intervention system for patients with nephrotic syndrome. Background Technology

[0002] In the clinical treatment of nephrotic syndrome, diuresis is a commonly used intervention aimed at promoting the excretion of excess water from the body. However, rapid fluid loss can lead to a reduction in intravascular fluid, resulting in blood concentration and a hypercoagulable state. In this hypercoagulable state, the microcirculation of local tissues deteriorates, and the tolerance of blood vessel walls to external pressure is significantly reduced. Therefore, the risk of venous thrombosis and tissue ischemia-necrosis in areas subjected to pressure during prolonged bed rest increases dramatically, and the physiological tolerance time is often shorter than the conventionally set safe turning interval.

[0003] Existing smart mattress monitoring systems primarily rely on the spatial displacement of the overall pressure center of gravity to determine changes in patient position. Because patients under pressure are extremely weak, their torsos are easily affected by gravity and may involuntarily slide downwards on the bed. This downward sliding causes a significant shift in the mattress's overall pressure center of gravity, leading the monitoring system to misinterpret it as a valid change in pressure points, resulting in incorrect turning decisions. Furthermore, hospital nursing systems typically generate turning schedules based on fixed countdown times, failing to dynamically adjust according to the patient's blood concentration caused by fluid metabolism. This makes routine nursing schedules severely lagging behind the actual physiological decline rate of patients in a hypercoagulable state. Summary of the Invention

[0004] To address the shortcomings of existing smart mattress monitoring systems, such as slippage misjudgment, and the rigidity of turning schedule instructions in hospital nursing systems, this invention aims to provide a complication risk early warning and nursing intervention system for patients with nephrotic syndrome. The specific technical solution adopted is as follows: This invention proposes a complication risk warning and nursing intervention system for patients with nephrotic syndrome, the system comprising: The data preprocessing module is used to obtain the hemoconcentration risk multiplier for the current monitoring period based on the target patient's serum albumin test value and historical fluid delivery data; and to construct a set of local high-pressure point coordinates for each monitoring period based on the pressure array matrix of the target patient in the current and historical monitoring periods of the smart mattress monitoring system. The turning-over judgment module is used to obtain the optimal matching point and pairing distance of the local high-pressure point in the current monitoring cycle based on the coordinate difference between the local high-pressure point coordinate set in the historical monitoring cycle of the last effective turning-over of the target patient and the local high-pressure point coordinate set in the current monitoring cycle; using the optimal matching point and pairing distance to filter the local high-pressure points in the historical monitoring cycle, the number of changed high-pressure points is obtained to obtain the local high-pressure point change ratio in the current monitoring cycle, and to analyze whether the target patient has undergone effective turning-over. The results analysis module is used to analyze the degree of pressure risk in the current monitoring cycle based on the change ratio of the local high pressure point and the blood concentration risk multiplier if no effective turning occurs. It also determines the risk warning result and nursing intervention instructions by combining the cumulative pressure risk of the previous monitoring cycle adjacent to the current monitoring cycle.

[0005] Furthermore, the method of obtaining the hemoconcentration risk multiplier for the current monitoring period based on the target patient's serum albumin test value and historical fluid delivery data includes: The historical liquid delivery data includes records of single liquid input volume and single liquid output volume; Based on the differences between all single fluid input and single fluid output records within the preset historical backtracking window during the current monitoring period, the net fluid transfer volume for the current monitoring period is obtained. Based on the target patient's latest serum albumin test value and the net volume of body fluid transfer, the corresponding risk multiplier is obtained from a preset numerical mapping table; The risk multiplier and the historically changed blood concentration risk multiplier are weighted and calculated according to a preset smoothing weight coefficient to obtain the blood concentration risk multiplier for the current monitoring period. The historically changed blood concentration risk multiplier is the blood concentration risk multiplier that is different from the value of the current monitoring period and is the time nearest neighbor.

[0006] Furthermore, the pressure array matrix of the target patient within the current monitoring period and historical monitoring periods of the intelligent mattress monitoring system constructs a set of local high-pressure point coordinates for each monitoring period, including: The pressure array matrix includes the real-time pressure value of each pressure point and its corresponding coordinates; For any given monitoring period, the pressure points in the pressure array matrix are divided into several connected regions, and the coordinates of the local high-pressure points in each connected region are calculated based on the real-time pressure value of each pressure point and its corresponding coordinates. Construct a set of local high-pressure point coordinates based on the coordinates of local high-pressure points in all connected regions.

[0007] Furthermore, the step of calculating the coordinates of the local high-pressure point in each connected region based on the real-time pressure value and corresponding coordinates of each pressure point includes: For any connected region, sort all real-time pressure values ​​within the connected region from largest to smallest, and extract the real-time pressure values ​​according to the sorting order and a preset number of windows. Calculate the local mean center coordinates of all real-time pressure values ​​within a preset quantity window, and use them as the coordinates of local high-pressure points in the connected region.

[0008] Further, the step of obtaining the optimal matching point and pairing distance of the local high-pressure point in the current monitoring period based on the coordinate difference between the current monitoring period and the set of local high-pressure point coordinates in the historical monitoring period of the last effective turning of the target patient includes: The set of local high-voltage point coordinates in the historical monitoring period is used as the reference set of local high-voltage point coordinates, and the local high-voltage points in the set of local high-voltage point coordinates in the historical monitoring period are used as the reference local high-voltage points. The Hungarian bipartite graph matching algorithm is used to perform a one-to-one matching between the reference local high-pressure point coordinate set and the local high-pressure point coordinate set of the current monitoring period, so as to obtain the optimal matching point and pairing distance of the reference local high-pressure point in the local high-pressure point coordinate set of the current monitoring period.

[0009] Further, the method of using the optimal matching point and pairing distance to filter local high-voltage points in the historical monitoring period, and obtaining the number of changed high-voltage points in the result to obtain the local high-voltage point change ratio in the current monitoring period, includes: The benchmark local high-pressure points that have not been matched with the optimal matching point in the coordinate set of local high-pressure points in the current monitoring period are identified as isolated points; The degree of dispersion of the pairing distance and the distribution of the number of isolated points are used to determine whether the target patient has experienced overall slippage. If overall sliding occurs, the change ratio of local high-pressure points in the current monitoring cycle will be set to 0; if no overall sliding occurs, the benchmark local high-pressure points and isolated points with a pairing distance greater than the preset trunk sliding distance threshold will be selected as high-pressure points to be changed. The ratio of the total number of changed high-voltage points to the total number of baseline local high-voltage points is used as the local high-voltage point change ratio for the current monitoring period.

[0010] Further, determining whether the target patient has experienced overall slippage based on the dispersion of the pairing distance and the distribution of the number of isolated points includes: Calculate the dispersion of the pairing distance; use the ratio of the number of isolated points to the number of reference local high-pressure points as the isolated point ratio; If the degree of dispersion is less than a preset deformation threshold and the proportion of isolated points is less than a preset proportion threshold, then overall sliding occurs; otherwise, overall sliding does not occur.

[0011] Furthermore, the analysis of whether the target patient has effectively turned over includes: Within a preset continuous period window, if the change ratio of local high pressure points in all monitoring cycles within the preset continuous period window is greater than or equal to the preset effective turning threshold, then the target patient has effectively turned over; otherwise, the target patient has not effectively turned over.

[0012] Furthermore, if no effective turning occurs, the pressure risk level for the current monitoring period is analyzed based on the change ratio of the local high-pressure point and the blood concentration risk multiplier, including: The complement of the local high-pressure point change ratio is used as the local pressure coefficient; The product of the blood concentration risk multiplier and the local pressure coefficient is used as the pressure risk increment for the current monitoring period.

[0013] Furthermore, the process of determining risk warning results and nursing intervention instructions by combining the accumulated risk of stress in the previous monitoring cycle adjacent to the current monitoring cycle includes: The cumulative value of the pressure risk in the previous monitoring period adjacent to the current monitoring period is added to the increment of the pressure risk in the current monitoring period to obtain the cumulative value of the pressure risk in the current monitoring period. The cumulative value of the pressure risk in the current monitoring period is normalized to obtain the continuous pressure risk value in the current monitoring period. If the sustained pressure risk value is less than the preset risk threshold, the pressure risk experienced by the target patient in the current monitoring period is within the physiologically tolerable safe range, and the system does not send nursing intervention instructions to medical staff; if the sustained pressure risk value is greater than or equal to the preset risk threshold, the pressure risk experienced by the target patient in the current monitoring period reaches the physiological tolerance limit, and the system issues an urgent intervention instruction to medical staff.

[0014] The present invention has the following beneficial effects: This invention obtains the hemoconcentration risk multiplier for the current monitoring period based on the target patient's serum albumin test value and historical fluid delivery data, thus quantifying the target patient's hemoconcentration status. Based on the pressure array matrix of the target patient within the current and historical monitoring periods of the intelligent mattress monitoring system, a set of local high-pressure point coordinates is constructed for each monitoring period. The macroscopic body posture pressure state of the target patient is abstracted into a set of discrete, definite spatial geometric coordinate points. By comparing the coordinate offsets across different monitoring periods, precise quantitative analysis of the actual changes in body position is achieved.

[0015] Based on the coordinate differences between the current monitoring cycle and the historical monitoring cycle in which the target patient was last effectively turned, the optimal matching point and pairing distance of the local high-pressure points in the historical monitoring cycle are obtained in the current monitoring cycle. This avoids misjudgments about whether the target patient has been effectively turned due to downward slippage caused by gravity. The number of changed high-pressure points in the results of screening local high-pressure points in the historical monitoring cycle using the optimal matching point and pairing distance is used to obtain the local high-pressure point change ratio in the current monitoring cycle, reflecting the degree of pressure relief for the target patient and analyzing whether the target patient has been effectively turned. If no effective turning has occurred, the pressure risk level in the current monitoring cycle is analyzed based on the local high-pressure point change ratio and the blood concentration risk multiplier. This represents the incremental pressure risk borne by the target patient in the current monitoring cycle due to their unique blood state and the continuously pressure-bearing parts of their body. Combining the cumulative pressure risk of the previous monitoring cycle adjacent to the current monitoring cycle, which reflects the cumulative value of the continuous pressure risk borne by the target patient, risk warning results and nursing intervention instructions are determined, realizing a leap from rigid execution to on-demand response in nursing care. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a structural diagram of a complication risk warning and nursing intervention system for patients with nephrotic syndrome, provided as an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a complication risk warning and nursing intervention system for patients with nephrotic syndrome proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following description, in conjunction with the accompanying drawings, details a specific scheme for a complication risk warning and nursing intervention system for patients with nephrotic syndrome provided by the present invention.

[0021] Please see Figure 1 The diagram illustrates a structural diagram of a complication risk warning and nursing intervention system for patients with nephrotic syndrome according to an embodiment of the present invention. The system includes: a data preprocessing module 101, a turning judgment module 102, and a result analysis module 103.

[0022] The data preprocessing module 101 is used to obtain the hemoconcentration risk multiplier for the current monitoring period based on the serum albumin test value of the target patient and historical fluid delivery data; and to construct a set of local high pressure point coordinates for each monitoring period based on the pressure array matrix of the target patient in the current and historical monitoring periods of the smart mattress monitoring system.

[0023] Serum albumin is a key indicator for assessing plasma colloid osmotic pressure. The lower the serum albumin level, the worse the blood vessels' ability to retain water. Even if the total fluid volume is not small, the effective circulating blood volume may be insufficient, making hemoconcentration more likely.

[0024] Historical fluid delivery data includes records of single fluid intake and output. Single fluid intake records refer to the specific milliliters of all fluids entering the patient's body through various routes, such as intravenous infusions, oral drinking, and tube feeding. Single fluid output records refer to the specific milliliters of all fluids expelled from the patient's body, such as urine and vomit. Historical fluid delivery data allows for the assessment of a patient's fluid balance.

[0025] Because patients with nephrotic syndrome typically experience significant fluid loss into subcutaneous tissue, leading to trunk edema, monitoring absolute body weight alone cannot accurately reflect the extent of intravascular blood dehydration. Therefore, combining serum albumin levels with historical fluid delivery data to analyze intravascular blood status allows for a more accurate assessment of the target patient's current true state of hemoconcentration, yielding a hemoconcentration risk multiplier that quantifies the target patient's hemoconcentration status.

[0026] In this embodiment of the invention, the system continuously monitors the medical record status update events of the target patient in the laboratory information system and the nursing information system. When a new test report data is generated in the laboratory information system or a new fluid delivery record is detected in the nursing information system, the serum albumin test value in the new test report data in the laboratory information system or the fluid delivery data in the new fluid delivery record in the nursing information system is immediately extracted.

[0027] A fixed monitoring cycle is set in the smart mattress monitoring system. A pressure array matrix representing the pressure state of the target patient is collected in real time during each monitoring cycle via a sensor array within the smart mattress. Each pressure point in the pressure array matrix corresponds to an independent sensing unit on the mattress, including the real-time pressure value and its corresponding coordinates in a two-dimensional bed surface coordinate system. The magnitude of the real-time pressure value reflects the real-time physical pressure exerted at that location. The two-dimensional bed surface coordinate system is directly mapped from the physical array layout of the pressure sensing units inside the smart mattress, with a physical spacing of 1 cm between adjacent sensor units, and the coordinate system unit is a physical length unit.

[0028] Specifically, the sensor unit at a fixed corner of the smart mattress (e.g., the top left corner) is used as the origin of the coordinate system (0, 0). A horizontal X-axis is established along the width of the mattress, and a vertical Y-axis is established along its length. The column index of each sensor unit in the pressure array matrix corresponds to its X-coordinate in the coordinate system, and the row index corresponds to its Y-coordinate. Thus, the matrix data can be directly correlated one-to-one with the two-dimensional spatial position of the physical bed surface.

[0029] Based on the pressure array matrix of each monitoring cycle of the intelligent mattress monitoring system, a set of local high-pressure point coordinates is constructed. The aim is to abstract the macroscopic body posture pressure state of the target patient into a set of discrete and definite spatial geometric coordinate points. By comparing the coordinate offsets of different monitoring cycles, a precise quantitative analysis of the actual changes in body position can be achieved.

[0030] In one specific implementation of this invention, the fixed monitoring period is set to 1 minute.

[0031] In this embodiment of the invention, if the target patient is being transferred to a hospital bed for the first time or the system is being powered on for the first time, the hospital's laboratory information system and nursing information system typically do not have the target patient's medical record status, making it impossible to determine the target patient's hemoconcentration risk multiplier based on the target patient's serum albumin test value and historical fluid delivery data. In this case, a preset physiological baseline constant is used as the target patient's initial hemoconcentration risk multiplier. This preset physiological baseline constant is a safe value that characterizes the patient's normal metabolic state.

[0032] In one specific implementation of this invention, the preset physiological baseline constant is set to 1.

[0033] In some implementations of this invention, low-intensity background noise inevitably exists in the matrix due to measurement errors in the mattress material or wrinkles caused by the patient's slight movement of the sheets. Without filtering, this noise will be misidentified as pressure points. Therefore, before constructing the set of local high-pressure point coordinates, a full matrix traversal is performed on the pressure array matrix for each monitoring cycle using a preset empty bed noise threshold. Pressure points with real-time pressure values ​​less than or equal to the preset empty bed noise threshold are identified as invalid pressure points, and their corresponding real-time pressure values ​​are forcibly cleared to zero to eliminate interference from background noise in subsequent coordinate extraction. Pressure points with real-time pressure values ​​greater than the preset empty bed noise threshold are identified as pressure points bearing the actual contact pressure of the patient's torso, and their corresponding real-time pressure values ​​and their row and column coordinates in the matrix are retained.

[0034] In one specific implementation of this invention, the preset empty bed noise threshold is the average ineffective pressure peak value obtained through multiple calibrations in an empty bed state. In this embodiment, the preset empty bed noise threshold is calculated to be 1.5 kPa.

[0035] In some implementations of this invention, all the data collected above and all the parameter data obtained in subsequent steps are stored in the global variable area. The global variable area is a shared data storage area in the system of this invention, used to persistently save key parameters.

[0036] The turning-over judgment module 102 is used to obtain the optimal matching point and pairing distance of the local high-pressure point in the current monitoring cycle based on the coordinate difference between the local high-pressure point coordinate set in the historical monitoring cycle of the last effective turning-over of the target patient and the local high-pressure point coordinate set in the current monitoring cycle; and to obtain the number of local high-pressure points that have changed in the result of screening the local high-pressure points in the historical monitoring cycle using the optimal matching point and pairing distance, thereby obtaining the local high-pressure point change ratio in the current monitoring cycle and analyzing whether the target patient has undergone effective turning-over.

[0037] Because patients with nephrotic syndrome who are bedridden for extended periods are highly susceptible to involuntary downward sliding on the bed due to gravity, traditional methods assess positional changes by calculating the overall pressure center displacement between the coordinates of local high-pressure points in the current monitoring cycle and the patient's last effective turning. In this case, the distance the patient slides downward due to gravity is incorrectly equated to the amount of center of gravity displacement, and is misjudged as an effective turning. This conventional method cannot distinguish whether the patient's downward sliding is due to gravity or an effective turning maneuver.

[0038] Therefore, to avoid misjudging whether a patient with nephrotic syndrome has been effectively turned over due to downward sliding caused by gravity, this embodiment obtains the optimal matching point and pairing distance of the local high-pressure point in the current monitoring period by comparing the coordinate difference between the local high-pressure point coordinate set of the current monitoring period and the historical monitoring period of the last effective turning of the target patient. For each local high-pressure point in the historical monitoring period, a unique local high-pressure point is found in the current monitoring period as the optimal matching point. The individual movement trajectory of each pressure point is tracked to analyze whether the pressure on the pressure area (such as the back) has been relieved.

[0039] For any local high-pressure point within the historical monitoring cycle of the target patient's last effective turning, if an optimal matching point exists within the current monitoring cycle and the matching distance is short, it indicates that the corresponding pressure point has only experienced a slight positional drift, and the corresponding pressure area has only slid downward due to gravity, without the pressure being relieved. If no optimal matching point exists within the current monitoring cycle or the matching distance is extremely far, it indicates that the corresponding pressure point has disappeared or experienced a long-distance jump, which means that the corresponding pressure area has left the bed surface and the pressure has been relieved. The percentage of localized high-pressure point changes represents the proportion of pressure-bearing points that have been relieved. It reflects the percentage of the target patient's body that, after excluding interference from downward slippage due to gravity, has actually moved away from its original pressure-bearing position on the mattress. A higher value indicates that more pressure-bearing areas have moved off the bed surface, and the greater the degree of pressure relief for the target patient. By analyzing the percentage of localized high-pressure point changes, it is possible to accurately determine whether the target patient has effectively turned over within the current monitoring period.

[0040] In this embodiment of the invention, if the target patient is lying in the hospital bed for the first time, the intelligent mattress monitoring system does not have a set of local high-pressure point coordinates for the historical monitoring period of the target patient's effective turning over. In this case, the set of local high-pressure point coordinates for the first monitoring period of the target patient is used as the set of local high-pressure point coordinates for the first effective turning over, and is used as a reference for subsequent monitoring periods.

[0041] The results analysis module 103 is used to analyze the degree of pressure risk in the current monitoring cycle based on the change ratio of the local high pressure point and the blood concentration risk multiplier if no effective turning occurs. It also determines the risk warning result and nursing intervention instructions by combining the cumulative pressure risk of the previous monitoring cycle adjacent to the current monitoring cycle.

[0042] The physiological tolerance limits of patients with nephrotic syndrome are dynamically influenced by individual factors such as their hemoconcentration state and whether the pressure points are effectively relieved. Traditional nursing instructions (such as turning patients every 2 hours) are static, general rules based on group experience. This fundamental mismatch between dynamic individual physiology and static, general instructions results in a severe delay in meeting the actual care needs of high-risk patients.

[0043] To resolve this contradiction, this invention analyzes the pressure risk level of the target patient based on the blood concentration risk multiplier (quantifying individual physiological vulnerability) and the proportion of changes in local high-pressure points during the current monitoring period (quantifying the effectiveness of pressure relief). This analysis represents the incremental pressure risk experienced by the target patient during the current monitoring period due to their unique blood state and the continuously compressed areas of their body. A higher pressure risk level indicates a higher risk of tissue ischemia and thrombosis for the target patient during the current monitoring period, and a faster rate of physiological depletion.

[0044] The cumulative stress risk from the previous monitoring cycle adjacent to the current monitoring cycle represents the cumulative stress risk the patient has continuously endured since the last effective turning of the target patient, reflecting the accumulation process of risk. Analyzing the stress risk level of the current monitoring cycle in conjunction with the cumulative stress risk from the previous monitoring cycle allows for accurate assessment of whether the target patient's current accumulated stress risk has exceeded their individual physiological tolerance level. This enables risk assessment and early warning, allowing for proactive nursing intervention when necessary, thus achieving a shift from rigid execution to on-demand response in nursing care.

[0045] In this embodiment of the invention, the system allocates and initializes a memory called a risk accumulator in the memory space. During the first monitoring cycle, the risk accumulator is initialized to 0. As the pressure risk increments in subsequent monitoring cycles, it is continuously accumulated and saved as the latest accumulated pressure risk value for the monitoring cycle. If the turning judgment module 102 determines that the target patient has effectively turned over in the current monitoring cycle, the accumulated pressure risk value of the previous monitoring cycle adjacent to the current monitoring cycle is cleared to zero, i.e., a forced zeroing operation of the risk accumulator is triggered.

[0046] In summary, this invention obtains the hemoconcentration risk multiplier for the current monitoring period based on the target patient's serum albumin test value and historical fluid delivery data. Based on the pressure array matrix of the target patient within the current and historical monitoring periods of the intelligent mattress monitoring system, a set of local high-pressure point coordinates is constructed for each monitoring period. According to the coordinate difference between the local high-pressure point coordinates of the current monitoring period and the historical monitoring period in which the target patient was last effectively turned, the optimal matching point and pairing distance of the local high-pressure points in the historical monitoring period are obtained in the current monitoring period, avoiding misjudgments of whether the target patient has been effectively turned due to downward sliding caused by gravity. The number of changed high-pressure points in the results of filtering local high-pressure points in the historical monitoring period using the optimal matching point and pairing distance is used to obtain the local high-pressure point change ratio for the current monitoring period, reflecting the degree of pressure relief for the target patient and analyzing whether the target patient has been effectively turned. If no effective turning occurs, the pressure risk level of the current monitoring cycle is analyzed based on the change ratio of the local high pressure point and the blood concentration risk multiplier. Combined with the cumulative pressure risk of the previous monitoring cycle adjacent to the current monitoring cycle, the risk warning result and nursing intervention instructions are determined, realizing the leap from rigid execution to on-demand response in nursing.

[0047] Preferably, in some implementations of the embodiments of the present invention, the blood concentration risk multiplier for the current monitoring period is obtained based on the serum albumin test value of the target patient and historical fluid delivery data, including: As can be seen from the data preprocessing module 101 above, the historical liquid delivery data includes records of single liquid input volume and records of single liquid output volume.

[0048] During the current monitoring period, the most recent laboratory test data of the target patient is collected in real time from the laboratory information system, and the serum albumin test value therein is used as the latest serum albumin test value of the target patient.

[0049] Based on the current monitoring period, extract all single fluid input and output records for the target patient from the nursing information system within a preset historical review window. Sum the individual milliliter values ​​for all single fluid input records and subtract the sum of the individual milliliter values ​​for all single fluid output records to obtain the net fluid transfer volume for the current monitoring period. If the net fluid transfer volume for the current monitoring period is negative, it indicates that the target patient is in a state of net dehydration with negative fluid balance; if the net fluid transfer volume for the current monitoring period is positive, it indicates that the patient is in a state of fluid retention with positive fluid balance.

[0050] In one specific implementation of this invention, the preset historical backtracking window is set to 24 hours.

[0051] Because serum albumin levels (biochemical concentration, g / L) and net fluid transfer volume (volume, mL) during the current monitoring period differ significantly in magnitude and physical dimensions, these two data points cannot be directly used in risk calculations. Therefore, this embodiment uses a pre-defined numerical mapping table to convert these two dimensions of indicators into a unified dimensionless risk multiplier.

[0052] Specifically, a numerical mapping table is pre-set in the local database. This table records different albumin concentration ranges, different net fluid transfer ranges, and preset coefficient values ​​corresponding to the cross-combinations of the two. The latest serum albumin test value and the net fluid transfer volume of the current monitoring period are combined into a binary variable pair, which is input into the numerical mapping table for coordinate matching to retrieve the specific numerical range containing this binary variable pair. The preset coefficient value mapped to the specific numerical range is extracted, and this preset coefficient value is used as the risk multiplier. In terms of mapping rules, the smaller the latest serum albumin test value and the larger the negative value of the net fluid transfer volume of the current monitoring period, the more it indicates that the patient's intravascular colloid osmotic pressure is extremely low and the body fluid loss is severe (i.e., the blood is extremely viscous), and the larger the corresponding risk multiplier value.

[0053] In one specific implementation of this invention, the preset numerical mapping table is generated based on statistical analysis of historical dehydration and thrombosis cases in the target department and pre-calibrated by medical experts. Specifically, it can be divided according to the formula [preset coefficient value = 1.0 + serum albumin detection value coefficient increment + net fluid transfer volume coefficient increment], as shown in the example below: A baseline coefficient of 1.0 is set when the serum albumin detection value is 30 g / L and the net fluid transfer volume is 0; for every 5 g / L decrease in serum albumin detection value, the corresponding preset coefficient value increases by 0.2 on top of the baseline coefficient; for every 500 ml decrease in net fluid transfer volume (negative values ​​increase), the corresponding preset coefficient value increases by 0.1 on top of the baseline coefficient. If the serum albumin detection value is greater than 30 g / L or the net fluid transfer volume is greater than 0, the corresponding preset coefficient value is the baseline coefficient.

[0054] Furthermore, to avoid abrupt changes in the risk multiplier obtained in the current monitoring period due to a single extreme test result or occasional errors in fluid delivery records, which could lead to a collapse in high-frequency monitoring calculations, this embodiment performs a smoothing weighting operation on the obtained risk multiplier. The risk multiplier and historically changed hemoconcentration risk multipliers are weighted according to a preset smoothing weighting coefficient to obtain the hemoconcentration risk multiplier for the current monitoring period. The historically changed hemoconcentration risk multiplier is the hemoconcentration risk multiplier with a value different from the current monitoring period and which is the closest in time neighbor. This ensures that even if the test data reflects a drastic change in the patient's vital signs, the updated hemoconcentration risk multiplier will transition with a gentle gradient, avoiding computational instability caused by high-frequency program calls.

[0055] Because the monitoring cycle of the smart mattress monitoring system is inconsistent with the serum albumin test value and historical fluid delivery data of the target patient in terms of time, there may be multiple monitoring cycles with the same blood concentration risk multiplier. Therefore, the blood concentration risk multiplier with the closest time neighbor and different historical changes is selected as a factor in calculating the blood concentration risk multiplier of the current monitoring cycle.

[0056] Specifically, the formula for obtaining the hemoconcentration risk multiplier is:

[0057] Where t represents the t-th monitoring period; The hemoconcentration risk multiplier represents the hemoconcentration risk multiplier for the t-th monitoring period; It is a preset smoothing weight coefficient, representing the weight of the risk multiplier in the t-th monitoring period; The risk multiplier represents the risk multiplier for the t-th monitoring period; The weight of the hemoconcentration risk multiplier representing historical changes; Blood concentration risk multiplier representing historical changes.

[0058] In one specific implementation of this invention, the preset smoothing weight coefficient is set to 0.3.

[0059] Preferably, in some implementations of the embodiments of the present invention, based on the pressure array matrix of the target patient within the current monitoring period and historical monitoring periods of the intelligent mattress monitoring system, a set of local high-pressure point coordinates for each monitoring period is constructed, including: When a person lies flat on a bed, the main pressure-bearing areas, such as the back or buttocks, are in large, continuous, flexible contact with the mattress. Under this flexible contact surface, the highest pressure area typically appears as a gentle peak, meaning that the real-time pressure values ​​at pressure points corresponding to multiple adjacent sensing units are similar or completely equal. To obtain stable local high-pressure points within the pressure array matrix for the current monitoring period, this implementation step locates these local high-pressure points by identifying connected regions within the pressure array matrix and calculating the coordinates of the local mean center of those regions.

[0060] As can be seen from the data preprocessing module 101 above, the pressure array matrix includes the real-time pressure value of each pressure point and its corresponding coordinates.

[0061] For any given monitoring period, using an image connected component marking algorithm (such as the commonly used two-pass scanning method or connected component tracing algorithm), the pressure points in the pressure array matrix of that monitoring period are divided into several connected regions. Based on the real-time pressure value of each pressure point and its corresponding coordinates, the coordinates of the local high-pressure point in each connected region are calculated.

[0062] The coordinates of local high-pressure points in all connected regions are packaged into a single coordinate data set to obtain the local high-pressure point coordinate set. This set accurately represents the instantaneous core trunk pressure distribution of the target patient within the corresponding monitoring period.

[0063] In one specific implementation of this invention, when performing an image connected component labeling algorithm on the pressure array matrix, pressure points whose real-time pressure values ​​are forcibly cleared to zero in the data preprocessing module 101 are considered background points, and pressure points with real-time pressure values ​​greater than 0 are considered foreground points. The specific steps are as follows: 1. Traverse the entire pressure array matrix in row and column order, and find the foreground pressure points that have not yet been assigned region labels as the starting point for the search.

[0064] 2. Assign a new connected region label to the starting point and check the pressure points in the 8 (or 4) adjacent directions around it.

[0065] 3. If adjacent points are also foreground pressure points and have not yet been labeled, then include them in the current connected region and assign them the same region label.

[0066] 4. Continue in this manner, recursively or iteratively searching for all adjacent foreground pressure points until the connected region can no longer be expanded outwards.

[0067] 5. Continue traversing the matrix to find the next unlabeled foreground pressure point, assign a new label, and repeat the above search expansion process until all foreground pressure points in the matrix have been assigned region labels, thereby dividing the matrix into several independent connected regions.

[0068] Furthermore, based on the real-time pressure value and corresponding coordinates of each pressure point, the coordinates of the local high-pressure points in each connected region are calculated, including: For any connected region, obtain all non-zero real-time pressure values ​​within the region and sort them from largest to smallest. Based on the sorting order and a preset quantity window, extract the corresponding number of real-time pressure values ​​and their corresponding coordinates. Calculate the arithmetic mean of all x-coordinates and y-coordinates of the extracted coordinates to obtain the x-coordinate and y-coordinate of the local mean center of the connected region, which serves as the coordinates of the local high-pressure point within the connected region.

[0069] In one specific implementation of this invention, the preset quantity window is set to 5. If there are fewer than 5 real-time pressure values ​​in the connected area, the coordinates of the maximum pressure value are used as the coordinates of the local high-pressure point in the connected area.

[0070] Preferably, in some implementations of the present invention, obtaining the optimal matching point and pairing distance of the local high-pressure point in the current monitoring period based on the coordinate difference between the current monitoring period and the coordinate set of the local high-pressure point in the historical monitoring period of the last effective turning of the target patient includes: Retrieve the set of local high-pressure point coordinates from the historical monitoring period of the last effective turning of the target patient from the global variable area. Use the set of local high-pressure point coordinates from the historical monitoring period as the reference set of local high-pressure point coordinates, and use the local high-pressure points within the set of local high-pressure point coordinates from the historical monitoring period as the reference local high-pressure points.

[0071] For any given reference local high-pressure point, calculate the coordinate distance between this reference local high-pressure point and all local high-pressure points within the coordinate set of local high-pressure points in the current monitoring period, in a two-dimensional bed surface coordinate system. It should be understood that the coordinate distance is the straight-line distance between two points in the two-dimensional bed surface coordinate system. The local high-pressure point ultimately paired with the reference local high-pressure point within the coordinate set of local high-pressure points in the current monitoring period is taken as the optimal matching point, and the coordinate distance between the reference local high-pressure point and the optimal matching point is taken as the pairing distance.

[0072] All coordinate distances obtained by calculating the coordinate set of all benchmark local high-voltage points in the current monitoring period are filled into a two-dimensional matrix to construct a distance cost matrix.

[0073] The Hungarian bipartite graph matching algorithm is used to solve the problem by minimizing the sum of the pairing distances corresponding to the optimal matching points of all benchmark local high-pressure points within the current monitoring period. A one-to-one matching process is performed between the set of benchmark local high-pressure point coordinates and the set of local high-pressure point coordinates for the current monitoring period to obtain the optimal matching point and pairing distance of the benchmark local high-pressure point within the set of local high-pressure point coordinates for the current monitoring period.

[0074] Specifically, the formula for calculating coordinate distance is:

[0075] in, Represents the first in the set of coordinates of the reference local high-pressure point Coordinates of a local high-pressure point; This represents the first local high-voltage point in the set of coordinates for the current monitoring period. Coordinates of a local high-pressure point; Indicates the first The first reference local high-pressure point and the first The coordinate distance of a local high-pressure point in the two-dimensional bed surface coordinate system; Indicates the first The abscissa of a reference local high-pressure point in a two-dimensional bed surface coordinate system; Indicates the first The ordinate of a reference local high-pressure point in a two-dimensional bed surface coordinate system; Indicates the first The abscissa of a local high-pressure point in a two-dimensional bed surface coordinate system; Indicates the first The ordinate of a local high-pressure point in the two-dimensional bed surface coordinate system.

[0076] Specifically, the steps for constructing the distance cost matrix are as follows: 1. Determine the matrix dimensions: The number of local high-pressure points m in the set of local high-pressure point coordinates of the statistical baseline and the number of local high-pressure points n in the set of local high-pressure point coordinates of the current monitoring period are taken as the maximum value N=max(m,n).

[0077] 2. Initialize an N*N square matrix: Create an empty square matrix with N rows and N columns. The rows represent the reference local high-pressure points in the reference local high-pressure point coordinate set, and the columns represent the local high-pressure points in the local high-pressure point coordinate set of the current monitoring period.

[0078] 3. Fill coordinate distance: Fill the coordinate distance between the i-th reference local high-pressure point in the reference local high-pressure point coordinate set and the j-th local high-pressure point in the local high-pressure point coordinate set of the current monitoring period into the i-th row and j-th column of the matrix.

[0079] 4. Fill empty nodes: Fill empty nodes with indices i>m or j>n in the matrix with a maximum value. This completes the construction of the distance cost matrix.

[0080] Specifically, the steps for obtaining the optimal matching point corresponding to the benchmark local high-pressure point using the Hungarian bipartite graph matching algorithm are as follows: 1. Input distance cost matrix: Input the N*N square matrix into the Hungarian bipartite graph matching algorithm module.

[0081] 2. Matrix row / column subtraction operation: Subtract the minimum value of each row of the matrix, and then subtract the minimum value of each column, so that there are potential matching positions with a value of 0 in each row and each column of the matrix.

[0082] 3. Covering zero elements: Cover all 0s in a matrix using the minimum number of horizontal or vertical lines.

[0083] 4. Adjust matrix coefficients: If the number of coverage lines is less than N, find the minimum value of the uncovered area, subtract the value from the uncovered area and add the value to the intersection area, until the number of coverage lines equals N.

[0084] 5. Determine the matching solution: Find N mutually independent 0s in the matrix (i.e., only one 0 in each row and column). The position of the 0 represents the optimal pairing relationship between the benchmark local high-pressure point coordinate set and the local high-pressure point coordinate set of the current monitoring period.

[0085] 6. Output results: Extract the coordinate distances corresponding to these optimal pairing positions as pairing distances, and output the unique optimal matching point and pairing distance of each benchmark local high-pressure point in the local high-pressure point coordinate set of the current monitoring period.

[0086] Preferably, in some implementations of the present invention, the number of changed high-voltage points in the result of screening local high-voltage points in the historical monitoring period using the optimal matching point and pairing distance is used to obtain the local high-voltage point change ratio for the current monitoring period, including: The benchmark local high pressure point that has not been matched with the optimal matching point in the local high pressure point coordinate set in the current monitoring period is determined as an isolated point. The degree of dispersion of the matching distance and the number distribution of isolated points are used to determine whether the target patient has experienced overall slippage.

[0087] If overall sliding occurs, the local high-pressure point change ratio for the current monitoring period is set to 0. If no overall sliding occurs, the baseline local high-pressure points matched to the optimal matching point are filtered according to the preset trunk sliding distance threshold. The trunk sliding distance threshold represents the maximum physical safety limit of downward sliding due to gravity that the patient's trunk can experience when not effectively turning over. For any baseline local high-pressure point matched to the optimal matching point, if the pairing distance is greater than the preset trunk sliding distance threshold, it is determined that the pressure-bearing part corresponding to the baseline local high-pressure point has undergone a significant displacement, exceeding the reasonable range of translation, and has actually moved away from the original mattress pressure position, thus relieving the pressure. This baseline local high-pressure point is then selected as a changed high-pressure point. If the pairing distance is less than or equal to the preset trunk sliding distance threshold, it is determined that the pressure-bearing part corresponding to the baseline local high-pressure point has slid downward due to gravity, and the pressure has not been relieved. This baseline local high-pressure point is then selected as an unchanged high-pressure point.

[0088] The ratio of the total number of changed high-voltage points to the total number of baseline local high-voltage points is used as the numerator, and the ratio of the total number of local high-voltage points to the baseline is used as the denominator. It should be understood that the total number of changed high-voltage points includes both baseline local high-voltage points filtered using a preset trunk sliding distance threshold and isolated points.

[0089] In one specific implementation of this invention, the preset torso sliding distance threshold is 10cm.

[0090] In this embodiment of the invention, if the target patient has completely left the bed, the real-time pressure values ​​within the pressure array matrix corresponding to the monitoring period are all 0, and there is no set of local high-pressure point coordinates. Therefore, if there is no set of local high-pressure point coordinates in either the current monitoring period or the historical monitoring period, the local high-pressure point change ratio for the current monitoring period is set to 1. If there is no set of local high-pressure point coordinates in either the current monitoring period or the historical monitoring period, it indicates that the patient has left the bed. To prevent the incremental pressure risk from continuing to accumulate, the local high-pressure point change ratio for the current monitoring period is set to 1.

[0091] Further, determining whether the target patient has experienced overall slippage based on the dispersion of the pairing distance and the distribution of the number of isolated points includes: In this embodiment of the invention, the variance of the pairing distance is calculated to represent the dispersion of the pairing distance. The ratio of the number of isolated points to the number of reference local high-pressure points is used as the isolated point ratio to represent the distribution of the number of isolated points.

[0092] If the variance is less than a preset deformation threshold and the proportion of isolated points is less than a preset proportion threshold, it indicates that the parallel displacement of each pressure point is rigid body sliding, and overall sliding occurs; if the variance is greater than or equal to the preset deformation threshold, or the proportion of isolated points is greater than or equal to the preset proportion threshold, it indicates that the body posture is reorganized and no overall sliding occurs.

[0093] In one specific implementation of this invention, under ideal overall sliding, the variance of the pairing distance should be close to 0. Considering that sensor noise and human body flexibility deformation can cause slight fluctuations, soft tissue traction and bed sheet deformation can produce relative displacement errors, and since the coordinate system unit in this embodiment is 1cm, a relative displacement deviation of 2cm to 3cm is allowed between each pressure point, i.e., the standard deviation of the deviation is 2 to 3, and the corresponding variance value is 4 to 9. In this embodiment, the preset deformation threshold is set to 5, and the preset proportional threshold is set to 0.2.

[0094] Preferably, in some implementations of the embodiments of the present invention, analyzing whether the target patient has effectively turned over includes: To prevent occasional discrete jumps in the sensor array (the real-time pressure value at a certain pressure point momentarily falls below the preset empty bed noise threshold) from causing the optimal matching point to disappear instantly, which in turn leads to misjudgments of drastic fluctuations in the local high-pressure point change ratio within a monitoring cycle, this embodiment of the invention introduces a continuous cycle confirmation logic.

[0095] Within a preset continuous period window, the relationship between the proportion of changes in local high-pressure points across all monitoring cycles within the continuous period window and the preset effective turning threshold is continuously compared. The preset effective turning threshold represents the minimum proportion of change required for the target patient to undergo a true and significant change in body position (i.e., effective turning).

[0096] If the change ratio of the local high-pressure points is greater than or equal to the preset effective turning judgment threshold, the target patient will make an effective turning, triggering the forced zeroing operation of the risk accumulator, and updating the local high-pressure point coordinate set of the current monitoring period to the new effective turning historical monitoring period local high-pressure point coordinate set as the benchmark for subsequent judgment.

[0097] If the change ratio of the local high pressure points cannot be greater than or equal to the preset effective turning threshold, then the target patient has not turned over effectively.

[0098] In one specific implementation of this invention, the preset continuous period window is set to 3 monitoring periods, and the preset effective turning over judgment threshold is set to 0.8.

[0099] Preferably, in some implementations of the present invention, if no effective turning occurs, the degree of pressure risk in the current monitoring period is analyzed based on the change ratio of the local high-pressure point and the blood concentration risk multiplier, including: The complement of the change ratio of local high pressure points is used as the local pressure coefficient, which indicates what proportion of the target patient's body has not been relieved of pressure during the current monitoring period.

[0100] The product of the hemoconcentration risk multiplier and the local pressure coefficient is used as the pressure risk increment for the current monitoring period. A larger pressure risk increment indicates a higher risk of tissue ischemia and thrombosis at the pressure site in the target patient.

[0101] Specifically, the formula for obtaining the increment of pressure risk is:

[0102] Where t represents the t-th monitoring period; This represents the increment of the pressure risk during the t-th monitoring period; The hemoconcentration risk multiplier represents the hemoconcentration risk multiplier for the t-th monitoring period; This represents the percentage change in local high-voltage points during the t-th monitoring period. This represents the local pressure coefficient during the t-th monitoring period.

[0103] Furthermore, based on the cumulative risk of stress in the previous monitoring cycle adjacent to the current monitoring cycle, risk warning results and nursing intervention instructions are determined, including: Extract the accumulated value of the pressure risk from the previous monitoring period adjacent to the current monitoring period from the risk accumulator, and add the accumulated value of the pressure risk to the increment of the pressure risk in the current monitoring period to obtain the accumulated value of the pressure risk in the current monitoring period.

[0104] Using the preset global experience maximum value as the maximum value and the preset global experience minimum value as the minimum value, the accumulated pressure risk value of the current monitoring period is normalized by the maximum and minimum values ​​to obtain the continuous pressure risk value of the current monitoring period. The range of the continuous pressure risk value is between 0 and 1.

[0105] It should be noted that if the cumulative value of the pressure risk in the current monitoring period is greater than the maximum value, the normalized value of the continuous pressure risk will be forced to be 1; if the cumulative value of the pressure risk in the current monitoring period is less than the minimum value, the normalized value of the continuous pressure risk will be forced to be 0.

[0106] In this embodiment of the invention, the global maximum and minimum empirical values ​​are pre-calibrated and written based on the statistical data of historical dehydration and thrombosis cases in the same department. The global maximum empirical value is set to 3.0, corresponding to extreme dehydration and hemoconcentration, and the global minimum empirical value is set to 0.5, corresponding to a safe state where the patient can move freely with significant changes in body position and metabolic indicators are normal.

[0107] If the sustained pressure risk value during the current monitoring period is less than the preset risk threshold, the pressure risk experienced by the target patient during the current monitoring period is within the physiologically tolerable safe range, and the system of this invention does not send nursing intervention instructions to medical staff. If the sustained pressure risk value during the current monitoring period is greater than or equal to the preset risk threshold, the pressure risk experienced by the target patient during the current monitoring period has reached the physiological tolerance limit, and the high risk of the target patient developing venous thrombosis has exceeded the limit. In this invention, the system issues an urgent intervention instruction to medical staff.

[0108] In one specific implementation of this invention, the preset risk threshold is set to 0.8.

[0109] In this embodiment of the invention, after issuing an expedited intervention command to medical staff, the system enters a continuous monitoring state. In each subsequent monitoring cycle, the system continuously reads the accumulated stress risk value for the corresponding monitoring cycle from the risk accumulator. When medical staff arrive at the bedside and actually reposition the target patient, a forced zeroing operation of the risk accumulator is triggered. Therefore, when the system reads that the accumulated stress risk value for the corresponding monitoring cycle in the risk accumulator is equal to zero in a subsequent monitoring cycle, it determines that the expedited intervention command has been successfully executed by the medical staff. Subsequently, a recovery command is sent to the medical order management system to revoke the expedited intervention command.

[0110] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A complication risk warning and nursing intervention system for patients with nephrotic syndrome, characterized in that, The system includes: The data preprocessing module is used to obtain the hemoconcentration risk multiplier for the current monitoring period based on the target patient's serum albumin test value and historical fluid delivery data; and to construct a set of local high-pressure point coordinates for each monitoring period based on the pressure array matrix of the target patient in the current and historical monitoring periods of the smart mattress monitoring system. The turning-over judgment module is used to obtain the optimal matching point and pairing distance of the local high-pressure point in the current monitoring cycle based on the coordinate difference between the local high-pressure point coordinate set in the historical monitoring cycle of the last effective turning-over of the target patient and the local high-pressure point coordinate set in the current monitoring cycle; using the optimal matching point and pairing distance to filter the local high-pressure points in the historical monitoring cycle, the number of changed high-pressure points is obtained to obtain the local high-pressure point change ratio in the current monitoring cycle, and to analyze whether the target patient has undergone effective turning-over. The results analysis module is used to analyze the degree of pressure risk in the current monitoring cycle based on the change ratio of the local high pressure point and the blood concentration risk multiplier if no effective turning occurs. It also determines the risk warning result and nursing intervention instructions by combining the cumulative pressure risk of the previous monitoring cycle adjacent to the current monitoring cycle.

2. The complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 1, characterized in that, The blood concentration risk multiplier for the current monitoring period is obtained based on the target patient's serum albumin test value and historical fluid delivery data, including: The historical liquid delivery data includes records of single liquid input volume and single liquid output volume; Based on the differences between all single fluid input and single fluid output records within the preset historical backtracking window during the current monitoring period, the net fluid transfer volume for the current monitoring period is obtained. Based on the target patient's latest serum albumin test value and the net volume of body fluid transfer, the corresponding risk multiplier is obtained from a preset numerical mapping table; The risk multiplier and the historically changed blood concentration risk multiplier are weighted and calculated according to a preset smoothing weight coefficient to obtain the blood concentration risk multiplier for the current monitoring period. The historically changed blood concentration risk multiplier is the blood concentration risk multiplier that is different from the value of the current monitoring period and is the time nearest neighbor.

3. The complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 1, characterized in that, The pressure array matrix of the target patient within the current and historical monitoring periods of the intelligent mattress monitoring system constructs a set of local high-pressure point coordinates for each monitoring period, including: The pressure array matrix includes the real-time pressure value of each pressure point and its corresponding coordinates; For any given monitoring period, the pressure points in the pressure array matrix are divided into several connected regions, and the coordinates of the local high-pressure points in each connected region are calculated based on the real-time pressure value of each pressure point and its corresponding coordinates. Construct a set of local high-pressure point coordinates based on the coordinates of local high-pressure points in all connected regions.

4. The complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 3, characterized in that, The calculation of the coordinates of the local high-pressure point in each connected region based on the real-time pressure value and corresponding coordinates of each pressure point includes: For any connected region, sort all real-time pressure values ​​within the connected region from largest to smallest, and extract the real-time pressure values ​​according to the sorting order and a preset number of windows. Calculate the local mean center coordinates of all real-time pressure values ​​within a preset quantity window, and use them as the coordinates of local high-pressure points in the connected region.

5. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 1, characterized in that, The step of obtaining the optimal matching point and pairing distance of the local high-pressure point in the current monitoring period based on the coordinate difference between the current monitoring period and the coordinate set of the local high-pressure point in the historical monitoring period of the last effective turning of the target patient includes: The set of local high-voltage point coordinates in the historical monitoring period is used as the reference set of local high-voltage point coordinates, and the local high-voltage points in the set of local high-voltage point coordinates in the historical monitoring period are used as the reference local high-voltage points. The Hungarian bipartite graph matching algorithm is used to perform a one-to-one matching between the reference local high-pressure point coordinate set and the local high-pressure point coordinate set of the current monitoring period, so as to obtain the optimal matching point and pairing distance of the reference local high-pressure point in the local high-pressure point coordinate set of the current monitoring period.

6. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 5, characterized in that, The method of filtering local high-voltage points in the historical monitoring period using the optimal matching point and pairing distance to obtain the number of changed high-voltage points and the proportion of local high-voltage point changes in the current monitoring period includes: The benchmark local high-pressure points that have not been matched with the optimal matching point in the coordinate set of local high-pressure points in the current monitoring period are identified as isolated points; The degree of dispersion of the pairing distance and the distribution of the number of isolated points are used to determine whether the target patient has experienced overall slippage. If overall sliding occurs, the change ratio of local high-pressure points in the current monitoring cycle will be set to 0; if no overall sliding occurs, the benchmark local high-pressure points and isolated points with a pairing distance greater than the preset trunk sliding distance threshold will be selected as high-pressure points to be changed. The ratio of the total number of changed high-voltage points to the total number of baseline local high-voltage points is used as the local high-voltage point change ratio for the current monitoring period.

7. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 6, characterized in that, The step of determining whether the target patient has experienced overall slippage based on the dispersion of the pairing distance and the distribution of the number of isolated points includes: Calculate the dispersion of the pairing distance; use the ratio of the number of isolated points to the number of reference local high-pressure points as the isolated point ratio; If the degree of dispersion is less than a preset deformation threshold and the proportion of isolated points is less than a preset proportion threshold, then overall sliding occurs; otherwise, overall sliding does not occur.

8. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 1, characterized in that, The analysis of whether the target patient has effectively turned over includes: Within a preset continuous period window, if the change ratio of local high pressure points in all monitoring cycles within the preset continuous period window is greater than or equal to the preset effective turning threshold, then the target patient has effectively turned over; otherwise, the target patient has not effectively turned over.

9. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 1, characterized in that, If no effective turning occurs, the pressure risk level for the current monitoring period is analyzed based on the change ratio of the local high-pressure point and the blood concentration risk multiplier, including: The complement of the local high-pressure point change ratio is used as the local pressure coefficient; The product of the blood concentration risk multiplier and the local pressure coefficient is used as the pressure risk increment for the current monitoring period.

10. A complication risk warning and nursing intervention system for patients with nephrotic syndrome according to claim 9, characterized in that, The process of combining the accumulated risk of stress in the previous monitoring cycle adjacent to the current monitoring cycle to determine the risk warning results and nursing intervention instructions includes: The cumulative value of the pressure risk in the previous monitoring period adjacent to the current monitoring period is added to the increment of the pressure risk in the current monitoring period to obtain the cumulative value of the pressure risk in the current monitoring period. The cumulative value of the pressure risk in the current monitoring period is normalized to obtain the continuous pressure risk value in the current monitoring period. If the sustained pressure risk value is less than the preset risk threshold, the pressure risk experienced by the target patient in the current monitoring period is within the physiologically tolerable safe range, and the system does not send nursing intervention instructions to medical staff; if the sustained pressure risk value is greater than or equal to the preset risk threshold, the pressure risk experienced by the target patient in the current monitoring period reaches the physiological tolerance limit, and the system issues an urgent intervention instruction to medical staff.

Images