Nurse station ward information collection method and system based on wireless network

Abstract

The embodiment of the invention provides a nurse station ward information collection method and system based on a wireless network. A flexible electronic skin network is integrated in a patient mattress, pressure and temperature data are captured through distributed piezoresistive units, a multi-modal thermodynamic diagram is generated, and data synchronization is achieved through a wireless ad hoc network protocol. And driving an air cushion bed actuator to perform gradient compensation according to a matching result of the pressure gradient track and the physiological parameters, transmitting a compressed hierarchical data packet to a nurse station gateway by adopting a remote wide area network protocol, reconstructing a three-dimensional pressure model and generating an early warning instruction. According to the technical scheme provided by the embodiment of the invention, the occurrence probability of health problems such as bedsore caused by long-time bed lying is remarkably reduced, and the nursing quality of the patient is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of wireless network technology, and more particularly to a method and system for collecting ward information at a nurse station based on a wireless network. Background Art

[0002] In the field of medical care, long-term bedridden patients are prone to complications such as pressure sores due to localized pressure concentration, which seriously affects the rehabilitation process and quality of life. Therefore, real-time monitoring of the patient's body pressure distribution and dynamic adjustment of the support pressure have become key needs of clinical care. Traditional nursing methods rely on manual timed turning, which is inefficient and difficult to achieve precise pressure regulation. With the development of wireless sensing technology and intelligent medical equipment, there is an urgent need for an intelligent solution that can collect patient body pressure data in real time, dynamically adjust support pressure, and achieve data synchronization and remote monitoring through wireless networks.

[0003] In the prior art, some solutions use a pressure sensor array embedded in the mattress to monitor body pressure distribution, and achieve pressure compensation through airbag adjustment of the air mattress. For example, some solutions collect body pressure data through distributed pressure sensors, and trigger the inflation and deflation operations of the air mattress in combination with preset thresholds. In addition, some solutions attempt to introduce wireless transmission technology to send the collected pressure data to the nurse station terminal for monitoring. These solutions have achieved the functions of body pressure monitoring and pressure regulation to a certain extent.

[0004] The existing solutions have the following major defects: First, the data collection and processing of the pressure sensor lacks multimodal fusion capabilities, and it is impossible to comprehensively evaluate the patient's body pressure distribution status by combining parameters such as temperature and impedance; second, the pressure compensation operation is mostly based on static threshold triggering, lacking the ability to dynamically adapt to the patient's physiological parameters (such as heart rate variability), resulting in insufficient compensation accuracy; finally, the existing wireless transmission solutions mostly use a single communication protocol, which is difficult to adapt to the high-reliability data transmission requirements in the complex wireless environment of the hospital, and lack the ability to encode the airbag deformation and body position displacement in time and space, resulting in the inability of the nurse station terminal to accurately reconstruct the three-dimensional pressure model. These problems limit the widespread application of existing technologies in clinical nursing and the improvement of their effects. Summary of the invention

[0005] The embodiment of the present application provides a method and system for collecting ward information at a nurse station based on a wireless network, so as to solve the problem in the prior art that the nurse station terminal cannot accurately reconstruct a three-dimensional pressure model.

[0006] In a first aspect, an embodiment of the present application provides a method for collecting ward information of a nurse station based on a wireless network, comprising:

[0007] A flexible electronic skin network is integrated inside the mattress, and the pressure distribution of the sacrum and coccyx is captured by a distributed piezoresistive unit. The impedance and temperature parameters are fused to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol.

[0008] Generate a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, drive the air mattress actuator to perform gradient compensation operation on the target airbag, and the trigger instruction forms a closed-loop link with the electronic skin through the wireless sensor network;

[0009] The airbag deformation parameters and body position deviation angles in the compensation operation are encoded in time and space, and hierarchical data packets are generated through compression algorithms. Dynamic channel allocation is used to transmit the data packets to the nurse station gateway via the remote wide area network protocol. The nurse station gateway is equipped with a multi-band aggregation module to adapt to the hospital wireless network.

[0010] The spatiotemporal coding in the hierarchical data packet is parsed at the nurse station terminal, and the three-dimensional pressure model of the airbag deformation and body position displacement is reconstructed based on the wireless transmission quality. The early warning instruction containing the pressure change rate and intervention duration is generated in combination with the vital sign file, and is mapped into a dynamic thermal topology map through the wireless broadcast mechanism.

[0011] Optionally, a trigger instruction is generated based on the matching of the pressure gradient trajectory in the multimodal thermal map with the physiological parameters to drive the air mattress actuator to perform a gradient compensation operation on the target airbag, and the trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network, including:

[0012] Based on the periodic morphological classification results of the sacral and coccygeal pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into a static load component and a dynamic fluctuation component, and a trigger instruction is generated by dynamically adjusting the pressure gradient determination threshold, wherein the trigger instruction includes a gradient compensation mode associated with the pressure distribution change rate;

[0013] According to the airbag basic compensation amount and compensation correction amount generated by the static load component and the dynamic fluctuation component, a dynamic mapping relationship between the target airbag pressure compensation parameter and the body support surface is established, and the air mattress actuator is driven to perform a gradient compensation operation through a gradient compensation parameter sequence, and the gradient compensation parameter sequence is spatially smoothed based on the time series characteristics of the dynamic fluctuation component;

[0014] During the gradient compensation operation, the physical position coordinates of the target airbag and the compensation parameter difference are interleaved in time sequence to generate a trigger instruction sequence including a three-dimensional compensation priority queue, and the trigger instruction sequence forms a closed-loop link through the wireless sensor network and the multi-node data synchronization of the electronic skin;

[0015] The gradient compensation parameter sequence and the trigger instruction sequence are dynamically coupled to form a composite instruction set, which synchronously includes the airbag deformation prediction parameters and the physiological response dynamic parameters, which are transmitted to the air bed actuator through the closed-loop link to perform the gradient compensation operation, and fed back to the nurse station gateway to reconstruct the pressure model.

[0016] Optionally, based on the periodic morphological classification results of the sacral and coccygeal pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into a static load component and a dynamic fluctuation component, and a trigger instruction is generated by dynamically adjusting the pressure gradient determination threshold. The trigger instruction includes a gradient compensation mode associated with the pressure distribution change rate, including:

[0017] Based on the periodic morphological classification result of the sacral and coccygeal pressure gradient trajectory, a waveform morphological matching algorithm is used to extract low-frequency baselines and decompose high-frequency disturbances of the continuous pressure waveform to generate a static load component and a dynamic fluctuation component. The static load component represents the steady-state characteristics of the patient's body pressure distribution, and the dynamic fluctuation component is associated with the heart rate variability data in the physiological parameters.

[0018] According to the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component, an adaptive adjustment mechanism for the pressure gradient determination threshold is established, and when the amplitude-frequency characteristic of the dynamic fluctuation component exceeds the adaptive adjustment threshold, a trigger instruction including a gradient compensation mode code is generated, and the gradient compensation mode code corresponds to the positive and negative polarity of the pressure distribution change rate;

[0019] The static load component is divided into multiple levels of support subdomains according to the anatomical region of the sacrum and coccyx, and the compensation weight of the static load component of each support subdomain is allocated in combination with the gradient compensation mode coding to generate a basic compensation amount bound to the physical position of the target airbag, and the basic compensation amount is superimposed with the compensation correction amount of the dynamic fluctuation component to form a gradient compensation parameter sequence;

[0020] The gradient compensation mode encoding is used to mark the gradient compensation parameter sequence with time priority, and a trigger instruction set including airbag deformation prediction parameters and compensation effective time window is generated. The trigger instruction set synchronously drives the air bed actuator through the wireless sensor network closed-loop link and feeds back to the nurse station gateway to update the three-dimensional pressure model.

[0021] Optionally, during the execution of the gradient compensation operation, the physical position coordinates of the target airbag and the compensation parameter difference are time-sequentially interleaved to generate a trigger instruction sequence including a three-dimensional compensation priority queue, and the trigger instruction sequence forms a closed-loop link through a wireless sensor network and multi-node data synchronization of the electronic skin, including:

[0022] Based on the physical position coordinates of the target airbag and the spatial distribution characteristics of the sacral pressure gradient trajectory, a position encoding strategy is used to generate a spatial topological relationship of the airbag group, wherein the spatial topological relationship includes radial distance parameters and axial angle parameters of adjacent airbags;

[0023] Performing time-series interleaving processing on the compensation parameter differences, calculating the attenuation gradient of the compensation amount between adjacent airbags based on the radial distance parameter in the spatial topological relationship, generating interpolation compensation parameters for smoothing the airbag deformation boundary, and superimposing the interpolation compensation parameters with the gradient compensation parameter sequence;

[0024] According to the superposition result of the interpolation compensation parameter and the gradient compensation parameter sequence, the compensation effective time window is phase aligned by a time interleaving algorithm to generate a three-dimensional compensation priority queue sorted by physical position coordinates, wherein each node in the queue contains a compensation amount, an attenuation gradient and a phase synchronization mark;

[0025] The three-dimensional compensation priority queue is dynamically matched with the multi-node pressure data collected in real time by the electronic skin, and the phase synchronization mark is calibrated through the wireless sensor network to generate a trigger instruction sequence containing the airbag deformation feedback parameters. The trigger instruction sequence synchronously drives the air mattress actuator through a closed-loop link and updates the three-dimensional pressure model of the nurse station gateway.

[0026] Optionally, an adaptive adjustment mechanism for the pressure gradient determination threshold is established according to the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component. When the amplitude-frequency characteristic of the dynamic fluctuation component exceeds the adaptive adjustment threshold, a trigger instruction including a gradient compensation mode code is generated. The gradient compensation mode code corresponds to the positive and negative polarity of the pressure distribution change rate, including:

[0027] Based on the phase synchronization relationship between the dynamic fluctuation component and the heart rate variability data, a physiological-pressure coupling strength parameter is generated through phase correlation analysis, and the physiological-pressure coupling strength parameter is used to construct an adaptive adjustment mechanism for the pressure gradient determination threshold;

[0028] The high-frequency energy distribution feature is extracted according to the amplitude-frequency characteristics of the dynamic fluctuation component, and combined with the dynamic offset calculation model in the adaptive adjustment mechanism, when the high-frequency energy distribution feature exceeds the adaptive adjustment threshold, a gradient compensation mode code bound to the positive and negative polarity of the pressure distribution change rate is generated, and the gradient compensation mode code is used as the core parameter of the trigger instruction;

[0029] The gradient compensation mode encoding is regionally mapped according to the anatomical distribution of the multi-level support subdomains of the sacral coccyx, and the compensation direction control parameters of each support subdomain are generated in combination with the spatial weight distribution result of the static load component. The compensation direction control parameters are used to correct the action direction of the basic compensation amount;

[0030] Based on the superposition relationship between the gradient compensation mode encoding and the compensation direction control parameter, a trigger instruction including the compensation effective direction and amplitude is generated. The trigger instruction synchronously drives the air bed actuator to perform directional gradient compensation through the wireless sensor network closed-loop link, and is fed back to the nurse station gateway to update the dynamic response parameters of the three-dimensional pressure model.

[0031] Optionally, the static load component is divided into multiple levels of support subdomains according to the anatomical region of the sacrum and coccyx, and the compensation weights of the static load components of each support subdomain are allocated in combination with the gradient compensation mode coding to generate a basic compensation amount bound to the physical position of the target airbag, and the basic compensation amount is superimposed with the compensation correction amount of the dynamic fluctuation component to form a gradient compensation parameter sequence, including:

[0032] Based on the physiological structure characteristics of the sacral anatomical area, a regional adaptive segmentation algorithm is used to divide the static load component into multi-level support subdomains, and a subdomain division result including partition contour parameters and pressure distribution density is generated. The partition contour parameters are bound to the physical position coordinates of the target airbag;

[0033] According to the positive and negative polarity characteristics encoded in the gradient compensation mode, a compensation weight dynamic allocation model of the support subdomain is established, and the initial compensation weight coefficient of each subdomain is generated in combination with the pressure distribution density. The initial compensation weight coefficient is dynamically attenuated with the compensation correction amount of the dynamic fluctuation component;

[0034] Performing spatial continuity constraint processing on the initial compensation weight coefficient, calculating a weight smoothing factor based on partition contour parameters of adjacent support subdomains, and generating a modified dynamic compensation weight coefficient, wherein the dynamic compensation weight coefficient is used to calculate a basic compensation amount bound to a target airbag;

[0035] The basic compensation amount and the compensation correction amount of the dynamic fluctuation component are vector-superimposed in the polarity direction of the gradient compensation mode encoding to generate a gradient compensation parameter sequence including the airbag position encoding and the weight attenuation parameter. The gradient compensation parameter sequence drives the air bed actuator through the trigger instruction set and synchronously updates the regional pressure distribution parameters of the three-dimensional pressure model in the nurse station gateway.

[0036] Optionally, the three-dimensional compensation priority queue is dynamically matched with the multi-node pressure data collected in real time by the electronic skin, the phase synchronization mark is calibrated through the wireless sensor network, and a trigger instruction sequence including the airbag deformation feedback parameter is generated. The trigger instruction sequence synchronously drives the air bed actuator through a closed-loop link and updates the three-dimensional pressure model of the nurse station gateway, including:

[0037] Based on the compensation amount, attenuation gradient and phase synchronization marking parameters in the three-dimensional compensation priority queue, a dynamic matching rule with the multi-node pressure data of the electronic skin is established, and a mapping relationship between the queue node and the pressure collection point is generated by measuring the similarity of the pressure distribution. The mapping relationship is used to extract the initial calibration value of the airbag deformation feedback parameter;

[0038] According to the timing offset between the phase synchronization mark parameter and the real-time pressure data of the electronic skin, a phase compensation algorithm is used to dynamically calibrate the effective time window of the trigger instruction sequence, and a modified trigger instruction sequence including the calibrated phase delay parameter is generated, wherein the modified trigger instruction sequence matches the response delay characteristic of the air mattress actuator;

[0039] Combined with the calibrated phase delay parameter and the airbag deformation feedback parameter in the correction trigger instruction sequence, the attenuation gradient and the compensation amount are dynamically adjusted through the feedback mechanism of the closed-loop link to generate an optimized trigger instruction sequence including the deformation compensation increment parameter, and the optimized trigger instruction sequence synchronously updates the node parameters of the three-dimensional compensation priority queue;

[0040] The optimized trigger instruction sequence is data-fused with the three-dimensional pressure model of the nurse station gateway, and the regional pressure distribution value of the pressure model is dynamically corrected based on the deformation compensation incremental parameter. The corrected pressure distribution value is fed back to the electronic skin network through a closed-loop link, forming real-time coordinated control of airbag deformation and pressure distribution.

[0041] In a second aspect, an embodiment of the present application provides a nurse station ward information collection system based on a wireless network, comprising:

[0042] A capture module is used to integrate a flexible electronic skin network inside a mattress, capture the sacral pressure distribution through a distributed piezoresistive unit, and fuse the impedance and temperature parameters to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol;

[0043] A driving module, for generating a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and matching the physiological parameters, driving the air mattress actuator to perform a gradient compensation operation on the target airbag, wherein the trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network;

[0044] The allocation module is used to perform spatiotemporal encoding on the airbag deformation parameters and the body position deviation angle in the compensation operation, generate hierarchical data packets through a compression algorithm, and transmit them to the nurse station gateway via a remote wide area network protocol using dynamic channel allocation. The nurse station gateway is configured with a multi-band aggregation module to adapt to the hospital wireless network;

[0045] The reconstruction module is used to parse the spatiotemporal coding in the hierarchical data packet at the nurse station terminal, reconstruct the three-dimensional pressure model of the airbag deformation and body position displacement based on the wireless transmission quality, generate early warning instructions containing the pressure change rate and intervention duration in combination with the vital sign file, and map it into a dynamic thermal topology map through the wireless broadcast mechanism.

[0046] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a wireless network-based nurse station ward information collection method as described in the first aspect above.

[0047] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a method for collecting ward information at a nurse station based on a wireless network as described in the first aspect.

[0048] In an embodiment of the present application, a flexible electronic skin network is integrated inside a patient mattress, and the pressure distribution of the sacral region is captured by a distributed piezoresistive unit, and the impedance and temperature parameters are fused to generate a multimodal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol; a trigger instruction is generated based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, and the air mattress actuator is driven to perform a gradient compensation operation on the target airbag. The trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network; the airbag deformation parameters and the body position offset angle in the compensation operation are spatiotemporally encoded, and a hierarchical data packet is generated through a compression algorithm. The data packet is transmitted to the nurse station gateway via a remote wide area network protocol using dynamic channel allocation. The nurse station gateway is configured with a multi-band aggregation module to adapt to the hospital wireless network; the spatiotemporal coding in the hierarchical data packet is parsed at the nurse station terminal, and a three-dimensional pressure model of airbag deformation and body position offset is reconstructed based on the wireless transmission quality. An early warning instruction containing the pressure change rate and intervention duration is generated in combination with the vital sign file, and is mapped into a dynamic thermal topology map through a wireless broadcast mechanism.

[0049] The technical solution of this application has the following beneficial effects:

[0050] By integrating a flexible electronic skin network inside the mattress, using a distributed piezoresistive unit to capture the sacral pressure distribution, and fusing the impedance and temperature parameters to generate a multimodal thermal map, the patient's body pressure distribution can be accurately monitored and intelligently adjusted. This method can not only synchronize data through the wireless ad hoc network protocol, but also drive the air mattress actuator to perform gradient compensation operations based on the pressure gradient trajectory and physiological parameter matching results in the multimodal thermal map, effectively improving nursing efficiency and patient comfort, while reducing the risk of complications caused by long-term bed rest, such as the incidence of health problems such as bedsores, thereby improving the overall quality of patient care.

[0051] Furthermore, the static load component and dynamic fluctuation component are generated by the periodic morphological classification of the sacral and coccygeal pressure gradient trajectory and the analysis of the heart rate variability data, and the trigger instruction is generated by dynamically adjusting the pressure gradient judgment threshold. The trigger instruction drives the air mattress actuator to implement the gradient compensation operation, forming a closed-loop link to ensure real-time adjustment and feedback of the pressure distribution. Through the dynamic matching of multimodal thermal images and physiological parameters, accurate monitoring and real-time adjustment of the patient's sacral and coccygeal pressure distribution are achieved, effectively reducing the risk of pressure sores; the formation of a closed-loop link ensures the timeliness and accuracy of the pressure compensation operation, improves nursing efficiency and patient comfort; the dynamic adjustment mechanism can adapt to the physiological characteristics of different patients, enhancing the universality and reliability of the system.

[0052] Furthermore, based on the phase synchronization relationship between the dynamic fluctuation component and the heart rate variability data, an adaptive adjustment mechanism for the pressure gradient judgment threshold is established. When the amplitude-frequency characteristics of the dynamic fluctuation component exceed the adaptive adjustment threshold, a trigger instruction containing the gradient compensation mode encoding is generated. The trigger instruction generates compensation direction control parameters through regional mapping and weight allocation, and finally drives the air mattress actuator to perform directional gradient compensation, and feeds back to the nurse station gateway to update the three-dimensional pressure model. Through the phase synchronization analysis of the dynamic fluctuation component and the heart rate variability data, accurate prediction and real-time response to pressure gradient changes are achieved, which significantly improves the accuracy and timeliness of pressure compensation; the adaptive adjustment mechanism can dynamically optimize the compensation strategy according to the patient's physiological state, enhancing the intelligence level of the system; the combination of directional gradient compensation and real-time update of the three-dimensional pressure model effectively improves the uniformity of pressure distribution and further reduces the risk of pressure sores.

[0053] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 A flowchart of a method for collecting ward information of a nurse station based on a wireless network provided by the present application is shown;

[0056] Figure 2 A schematic diagram of the structure of a nurse station ward information collection system based on a wireless network provided by the present application is shown;

[0057] Figure 3 A schematic diagram of the structure of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0060] This application proposes an intelligent pressure control method based on flexible electronic skin and closed-loop control to address the problems of insufficient multimodal data fusion, delayed compensation response, and low wireless transmission reliability in traditional bed pressure monitoring systems. By integrating a flexible sensor network to collect multimodal physiological data in real time, combining a dynamic gradient compensation algorithm to achieve precise pressure regulation, and using space-time coding and multi-band aggregation technology to optimize data transmission, a full-link closed-loop system from data collection to nurse station warning is finally constructed. This design realizes the organic synergy of body pressure distribution monitoring, real-time compensation, efficient transmission, and three-dimensional model reconstruction, solving the pain points of data isolation, delayed response, and unstable transmission in the existing technology.

[0061] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0062] Figure 1 A flowchart of a method for collecting ward information of a nurse station based on a wireless network is provided for an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0063] 101. A flexible electronic skin network is integrated inside a mattress, and the pressure distribution of the sacrum and coccyx is captured by a distributed piezoresistive unit, and impedance and temperature parameters are integrated to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol;

[0064] In this step, the flexible electronic skin network refers to the distributed piezoresistive sensor array embedded in the patient mattress, which is used to capture the pressure distribution data of the patient's sacrum and coccyx in real time. The impedance parameter reflects the state of tissue compression, and the temperature parameter is used to monitor changes in local blood circulation. The two are fused to generate a multimodal thermal map to intuitively display the coupled distribution of pressure and temperature. The wireless ad hoc network protocol realizes multi-node data synchronization, ensuring efficient networking and low-latency communication of the sensor network.

[0065] In the embodiment of the present application, the flexible electronic skin network is composed of multiple piezoresistive units and temperature sensors, which are deployed inside the mattress in a honeycomb topology. After the sensor nodes collect pressure, impedance and temperature data, they are aligned and normalized in time and space by the edge computing module to generate a multimodal thermal map. The wireless ad hoc network protocol adopts a dynamic routing algorithm to adaptively adjust the communication path between nodes according to the signal strength to ensure data synchronization accuracy and network stability.

[0066] A hospital has deployed a flexible electronic skin network in a patient mattress. The piezoresistive units are distributed in a 10cm×10cm grid, and each node is integrated with a temperature sensor. When the patient lies down, the sacrococcygeal pressure and temperature data are uploaded to the edge gateway in real time to generate a multimodal thermal map covering the entire sacrococcygeal area. When the pressure in a certain area exceeds the threshold and the temperature is abnormal, the system marks it as a high-risk area, triggering subsequent compensation operations.

[0067] 102. Generate a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, drive the air mattress actuator to perform gradient compensation operation on the target airbag, and the trigger instruction forms a closed-loop link with the electronic skin through the wireless sensor network;

[0068] In this step, the pressure gradient trajectory of the multimodal thermal map refers to the spatial variation trend of the pressure distribution, and the physiological parameters include vital sign data such as heart rate variability (HRV). The trigger command is dynamically generated by matching the pressure gradient with the physiological characteristics, driving the air mattress actuator to perform gradient compensation on the target airbag, forming a closed-loop link between the electronic skin sensor data and the actuator action.

[0069] In the embodiment of the present application, the pressure gradient trajectory is generated by calculating the spatiotemporal derivative of the thermal map and is correlated with the patient's real-time HRV data (such as RR interval). When an abnormal pressure gradient is detected and the HRV index fluctuates, the system generates a trigger instruction including compensation mode, action area and intensity, which is sent to the target airbag controller through the wireless sensor network to achieve dynamic adjustment of the inflation and deflation volume.

[0070] When the pressure gradient on the left side of the sacrum and coccyx of a patient suddenly rises in step 101 and HRV shows sympathetic nerve excitement, the system determines that this area needs to be decompressed first. The trigger command specifies that the left No. 3 airbag is deflated by 20% and the right No. 5 airbag is inflated by 15% to form pressure gradient compensation. After the actuator responds, the electronic skin network feeds back the compensation effect in real time, and the closed-loop link continuously optimizes the command parameters.

[0071] 103. Perform spatiotemporal encoding on the airbag deformation parameters and the body position deviation angle in the compensation operation, generate hierarchical data packets through a compression algorithm, and transmit them to the nurse station gateway through a remote wide area network protocol using dynamic channel allocation. The nurse station gateway is configured with a multi-band aggregation module to adapt to the hospital wireless network;

[0072] In this step, the airbag deformation parameters include physical quantities such as deformation amount and deformation rate, and the body position deviation angle is measured by inertial sensors to measure the patient's posture changes. Spatiotemporal coding marks the deformation and body position data by timestamp and spatial coordinates, hierarchical data packets reduce redundancy through compression algorithms, and dynamic channel allocation switches the transmission frequency band according to the hospital's wireless environment to improve reliability.

[0073] In the embodiment of the present application, the airbag deformation parameters and body position data are fused through Kalman filtering to generate a time-space coding sequence with a timestamp. The compression algorithm combines differential coding with Huffman coding to reduce the amount of data by more than 60%. The dynamic channel allocation module monitors the interference intensity between the hospital Wi-Fi and 5G frequency bands in real time, and gives priority to transmitting data packets in the frequency band with the highest signal-to-noise ratio.

[0074] After the No. 3 airbag is deflated in step 102, the system records its deformation as 8mm, and detects that the patient's body tilts 5° to the right. The spatiotemporal coding associates the deformation data with the body angle as "timestamp T1, position (X3, Y5), deformation -8mm, tilt 5° to the right". The compressed data packet is transmitted to the nurse station gateway via the 5G frequency band, and the gateway multi-band aggregation module automatically switches to the low-interference channel to complete the reception.

[0075] 104. Parse the spatiotemporal coding in the hierarchical data packet at the nurse station terminal, reconstruct the three-dimensional pressure model of the airbag deformation and body position displacement based on the wireless transmission quality, generate early warning instructions containing the pressure change rate and intervention duration in combination with the vital sign file, and map it into a dynamic thermal topology map through the wireless broadcast mechanism.

[0076] In this step, the three-dimensional pressure model reconstructs the spatial relationship between airbag deformation and body position displacement through the inverse kinematics algorithm, and the vital signs file integrates the patient's historical pressure data and medical history information. The early warning instructions are generated based on the pressure change rate and intervention duration thresholds, and the dynamic thermal topology map displays the evolution of pressure distribution in real time through color gradients and animation effects.

[0077] In the embodiment of the present application, after parsing the spatiotemporal coding, the nurse station terminal dynamically adjusts the model reconstruction weight in combination with the wireless transmission packet loss rate. The three-dimensional pressure model uses a finite element simulation algorithm to map the deformation parameters into a body pressure distribution cloud map. The early warning instruction is generated by comparing the current pressure change rate with a preset safety threshold (such as >5kPa / min) and is pushed to the nursing staff's mobile terminal via wireless broadcast.

[0078] After the data packet transmitted in step 103 is parsed at the nurse station terminal, the system detects that the pressure change rate of the patient's sacral coccyx reaches 6kPa / min (exceeding the safety threshold). The three-dimensional model shows that the pressure in this area is concentrated and continues to rise, and the early warning instruction recommends "adjusting the body position within 10 minutes". The dynamic thermal topology map marks the high-risk area with flashing red, and reminds the nurse to intervene through the ward broadcast.

[0079] In summary, in steps 101 to 104, the high-precision acquisition and fusion of multimodal data is achieved through the flexible electronic skin network, and the pressure regulation response speed is significantly improved in combination with the dynamic gradient compensation algorithm; the space-time coding and multi-band transmission technology ensure the integrity and real-time performance of data in complex wireless environments; the intelligent generation of three-dimensional pressure models and early warning instructions realizes the closed-loop management of the entire process from data monitoring to nursing decision-making. Ultimately, the incidence of pressure sores is reduced by more than 30%, the nursing response efficiency is increased by 50%, and the frequency of manual intervention is reduced, which has significant clinical value and economic benefits.

[0080] In order to solve the problems of response lag, insufficient compensation accuracy and lack of closed-loop feedback caused by static threshold triggering in existing pressure compensation schemes, this step realizes accurate gradient compensation from data analysis to actuator control by dynamically decomposing the pressure waveform, establishing dynamic mapping of compensation parameters, optimizing instruction priority through time-series interleaving, and multi-parameter fusion closed-loop feedback. This design deeply integrates physiological characteristics with dynamic changes in pressure, breaks through the limitations of traditional one-way compensation, and forms an adaptive closed-loop control link. In some embodiments, in step 102, a trigger instruction is generated based on the matching of the pressure gradient trajectory and physiological parameters in the multimodal thermal map to drive the air mattress actuator to perform gradient compensation operations on the target airbag. The trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network, including:

[0081] 1021. Based on the periodic morphological classification results of the sacral and coccygeal pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into a static load component and a dynamic fluctuation component, and a trigger instruction is generated by dynamically adjusting the pressure gradient determination threshold, wherein the trigger instruction includes a gradient compensation mode associated with the pressure distribution change rate;

[0082] The static load component refers to the slowly changing steady-state pressure component in the patient's body pressure distribution, reflecting the long-term pressure state; the dynamic fluctuation component refers to the rapid pressure fluctuation component caused by physiological activities (such as breathing and heart rate variability). The pressure gradient judgment threshold is dynamically adjusted according to the heart rate variability (HRV) data to identify abnormal pressure change rates. The gradient compensation mode code defines the compensation direction and intensity under different pressure change rates.

[0083] In the embodiment of the present application, a wavelet transform algorithm is used to perform multi-scale decomposition of the continuous pressure waveform, extracting the low frequency band (0.1-0.5Hz) as the static load component and the high frequency band (0.5-2Hz) as the dynamic fluctuation component. The judgment threshold is dynamically adjusted according to the low frequency / high frequency power ratio (LF / HF) in the HRV data: when LF / HF>2 (sympathetic nerves are active), the judgment threshold is reduced by 20% to sensitively capture abnormal pressure fluctuations. The gradient compensation mode encoding in the trigger instruction includes two types of logic: "positive compensation" (deflation when pressure increases) and "negative compensation" (inflation when pressure decreases).

[0084] 1022. According to the airbag basic compensation amount and compensation correction amount generated by the static load component and the dynamic fluctuation component, a dynamic mapping relationship between the target airbag pressure compensation parameter and the body support surface is established, and the air mattress actuator is driven to perform a gradient compensation operation through a gradient compensation parameter sequence, and the gradient compensation parameter sequence is spatially smoothed based on the time series characteristics of the dynamic fluctuation component;

[0085] The basic compensation of the airbag is calculated based on the spatial distribution of the static load component, and the compensation correction is generated by weighting the amplitude and frequency of the dynamic fluctuation component. The dynamic mapping relationship refers to associating the compensation parameters with the geometric characteristics of the patient's position support surface (such as curvature and contact area) to ensure that the compensation amount matches the body pressure distribution in space. Spatial smoothing eliminates local mutations by interpolating adjacent airbag compensation amounts.

[0086] In the embodiment of the present application, the basic compensation amount is allocated by region according to the pressure integral value of the static load component. For example, the area with a pressure integral value > 50 kPa·s is allocated a basic deflation amount of 10%. The compensation correction amount is calculated by multiplying the amplitude (peak-to-peak value) of the dynamic fluctuation component by the main frequency. For example, a correction amount of ±3% is generated when the amplitude is 5 kPa and the frequency is 1 Hz. The dynamic mapping uses a surface fitting algorithm to map the airbag position to the three-dimensional coordinates of the patient's body pressure support surface. The spatial smoothing process uses a Gaussian filter to perform a weighted average of the compensation amounts of adjacent 3×3 airbag groups.

[0087] 1023. During the gradient compensation operation, the physical position coordinates of the target airbag and the compensation parameter difference are interleaved in time sequence to generate a trigger instruction sequence including a three-dimensional compensation priority queue, and the trigger instruction sequence forms a closed-loop link through the wireless sensor network and the multi-node data synchronization of the electronic skin;

[0088] Time interleaving refers to sorting the compensation parameters of different time windows according to the physical position of the airbag and then cross-combining them to avoid command conflicts. The three-dimensional compensation priority queue is generated according to the rule of "spatial proximity first, compensation difference first", and the queue node contains the compensation amount, effective timestamp and spatial coordinate encoding. The closed-loop link verifies the compensation effect through the real-time pressure data of the electronic skin and dynamically corrects the queue parameters.

[0089] In the embodiment of the present application, the timing interleaving algorithm divides the compensation instruction into 10ms time slices and generates a priority index according to the XY axis value of the airbag position coordinate. For example, the airbag instruction priority of the sacral center area (X5, Y5) is higher than that of the edge area (X1, Y1). After the queue is generated, the electronic skin network collects the compensated pressure data in real time. If the compensation deviation of a certain area is detected to be >15%, the airbag instruction is reinserted into the queue head for priority execution.

[0090] 1024. Dynamically couple the gradient compensation parameter sequence and the trigger instruction sequence into a composite instruction set, wherein the composite instruction set synchronously includes the airbag deformation prediction parameters and the physiological response dynamic parameters, which are transmitted to the air mattress actuator through the closed-loop link to perform the gradient compensation operation, and fed back to the nurse station gateway to reconstruct the pressure model.

[0091] The composite instruction set aligns the gradient compensation parameter sequence with the trigger instruction sequence by timestamp, and integrates the airbag deformation prediction parameters (such as expected deformation amount and deformation rate) and physiological response parameters (such as HRV change trend). Pressure model reconstruction refers to updating the three-dimensional model of the nurse station according to the compensated pressure distribution, forming an "execution-feedback-optimization" closed loop.

[0092] In the embodiment of the present application, the composite instruction set is encapsulated in JSON format and includes the following fields: {"airbag ID": A3, "compensation amount": -20%, "deformation prediction": 8mm, "HRV trend": ↑15%}. After receiving the instruction, the nurse station gateway calculates the pressure distribution after compensation through finite element simulation. If the predicted pressure gradient still exceeds the threshold, a secondary compensation instruction is generated and fed back to the actuator.

[0093] Here is a specific example:

[0094] When a patient is lying in bed, the pressure gradient on the right side of the sacrum and coccyx continues to rise (static load component reaches 60kPa), and HRV shows LF / HF=2.3 (sympathetic nerves are active). In step 1021, the system determines that the threshold is reduced by 20%, and the dynamic fluctuation component detects abnormal fluctuations with an amplitude of 6kPa and a frequency of 1.2Hz, and generates a "negative compensation" instruction (inflation 15%). In step 1022, the basic compensation amount is allocated to inflate the right airbag by 10%, and the correction amount is added by 5%. After smoothing by Gaussian filtering, a gradient compensation parameter sequence is generated. In step 1023, the right central airbag (X7, Y5) instruction is inserted into the first position of the queue, and the actuator is driven to inflate 15% within 5ms. In step 1024, the composite instruction set synchronously predicts the deformation of 7mm and monitors the HRV down to LF / HF=1.8. The three-dimensional model of the nurse station shows that the pressure gradient has returned to normal, and the closed-loop link terminates compensation.

[0095] Through dynamic pressure decomposition and threshold adaptive adjustment (1021), abnormal pressure fluctuations can be accurately captured; compensation parameter generation based on spatial smoothing and dynamic mapping (1022) ensures uniformity of pressure regulation and ergonomic adaptation; timing interleaving and priority queue optimization (1023) significantly improve the efficiency and reliability of multi-airbag collaborative compensation; composite instruction set and closed-loop feedback mechanism (1024) form a complete control link from local compensation to global model update. Ultimately, the pressure compensation response time is shortened to less than 200ms, the compensation accuracy is improved by 40%, and the false trigger rate is reduced to less than 5%.

[0096] In order to further improve the accuracy and physiological adaptability of pressure gradient compensation, this solution constructs a multi-level collaborative control mechanism from data analysis to compensation execution through waveform decomposition, dynamic threshold adjustment, regional weight allocation and timing optimization marking. This design deeply integrates periodic pressure morphology with physiological characteristics to solve the problems of single compensation mode, uneven regional response and closed-loop update delay in traditional solutions, and achieve adaptive dynamic balance of patient body pressure distribution. In some embodiments, based on the periodic morphological classification results of the sacral pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into static load components and dynamic fluctuation components, and a trigger instruction is generated by dynamically adjusting the pressure gradient judgment threshold. The trigger instruction contains a gradient compensation mode associated with the pressure distribution change rate, including:

[0097] 2021. Based on the periodic morphological classification result of the sacral and coccygeal pressure gradient trajectory, a waveform morphological matching algorithm is used to perform low-frequency baseline extraction and high-frequency disturbance decomposition on the continuous pressure waveform to generate a static load component and a dynamic fluctuation component, wherein the static load component represents the steady-state characteristics of the patient's body pressure distribution, and the dynamic fluctuation component is associated with the heart rate variability data in the physiological parameters;

[0098] The low-frequency baseline refers to the slowly changing steady-state component (0.1-0.5Hz) in the pressure waveform, which represents the long-term stable state of the patient's body pressure distribution; the high-frequency disturbance refers to the fast-fluctuating component (0.5-2Hz) caused by physiological activities (such as breathing and heart rate variability). The waveform morphology matching algorithm recognizes the periodic characteristics of the actual waveform and separates the high- and low-frequency components by presetting typical pressure waveform templates (such as sine waves and square waves).

[0099] In the embodiment of the present application, the empirical mode decomposition (EMD) algorithm is used to adaptively decompose the continuous pressure waveform, extract the first three intrinsic mode functions (IMFs) as high-frequency disturbance components, and the remaining trend terms as low-frequency baselines. For example, after the sacrococcygeal pressure waveform of a patient is decomposed, IMF1 (1.5Hz) reflects the fluctuation of heart rate variability, IMF2 (0.8Hz) corresponds to the respiratory rhythm, and the low-frequency baseline shows that the steady-state pressure value is 45kPa.

[0100] 2022. Establish an adaptive adjustment mechanism for the pressure gradient determination threshold value according to the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component, and generate a trigger instruction including a gradient compensation mode code when the amplitude-frequency characteristic of the dynamic fluctuation component exceeds the adaptive adjustment threshold value, wherein the gradient compensation mode code corresponds to the positive and negative polarity of the pressure distribution change rate;

[0101] The phase synchronization relationship refers to the consistency of the phase difference between the dynamic fluctuation component and the heart rate variability (HRV) signal in the time and frequency domain. The phase difference accumulation value is calculated by the phase-locked loop (PLL) algorithm. The amplitude-frequency characteristics include the amplitude (peak-to-peak value) of the dynamic fluctuation component multiplied by the main frequency. The adaptive adjustment threshold is dynamically scaled according to the LF / HF ratio of HRV (sympathetic-vagal balance indicator). The gradient compensation mode encoding defines that deflation compensation is performed when the pressure change rate is positive (pressure rises), and inflation compensation is performed when the rate is negative.

[0102] In the embodiment of the present application, when the phase difference between the dynamic fluctuation component and the HRV signal is detected to be less than 15° and lasts for 3 cycles, it is determined to be phase synchronization. If the dynamic fluctuation amplitude-frequency characteristic (such as amplitude 5kPa×frequency 1.2Hz=6kPa·Hz) exceeds the threshold (threshold=4×LF / HF) at this time, the compensation mode code is generated. For example, when LF / HF=1.5, the threshold is 6kPa·Hz, and if the measured value is 7kPa·Hz, the "negative compensation" code (inflation 10%) is triggered.

[0103] 2023. Divide the static load component into multiple levels of support subdomains according to the anatomical region of the sacrum and coccyx, allocate compensation weights to the static load component of each support subdomain in combination with the gradient compensation mode coding, generate a basic compensation amount bound to the physical position of the target airbag, and superimpose the basic compensation amount with the compensation correction amount of the dynamic fluctuation component to form a gradient compensation parameter sequence;

[0104] The multi-level support subdomain refers to the pressure response area divided according to the anatomical structure of the sacrococcygeal region (such as the ischial tuberosity area and the sacral area). Each subdomain contains a group of associated airbag units. The compensation weight allocation is based on the positive and negative polarity encoded in the gradient compensation mode. The weight of the positive compensation (deflation) area increases, and the weight of the negative compensation (inflation) area decreases. The basic compensation amount is the product of the pressure integral value of the subdomain static load component and the weight.

[0105] In the embodiment of the present application, the Voronoi diagram algorithm is used to divide the sacrum and coccyx into 5 supporting subdomains, such as the ischial tuberosity subdomain (weight coefficient 1.2) and the central sacral subdomain (weight coefficient 1.0). If the pressure integral of the static load component of a subdomain is 50 kPa·s and the gradient compensation mode is negative (inflation), then the basic compensation amount = 50 × 1.0 = 50 mL of inflation, and the compensation correction amount is the amplitude-frequency value of the dynamic fluctuation component (such as 6 kPa·Hz) plus 10 mL, and the total compensation amount is 60 mL.

[0106] 2024. The gradient compensation parameter sequence is marked with time priority through the gradient compensation mode encoding to generate a trigger instruction set including airbag deformation prediction parameters and compensation effective time window. The trigger instruction set synchronously drives the air mattress actuator through the wireless sensor network closed-loop link and feeds back to the nurse station gateway to update the three-dimensional pressure model.

[0107] The timing priority mark is generated according to the urgency of the compensation effective time window (such as pressure change rate > 5kPa / min is marked as high priority) and the airbag spatial position (central area is prioritized). The airbag deformation prediction parameters are calculated through the deformation-pressure transfer function, and the compensation effective time window defines the start and end timestamps of the instruction execution.

[0108] In the embodiment of the present application, the trigger instruction set adopts the priority label "P1-P3", where the P1 level instruction (central area, change rate>8kPa / min) is executed within 10ms, and the P2 level instruction (edge ​​area, rate 5-8kPa / min) is executed within 50ms. The deformation prediction parameters are calculated by the linear regression model of the airbag deformation history data and the current compensation amount. For example, a compensation amount of 60mL predicts a deformation amount of 6mm. After receiving the instruction, the nurse station gateway updates the predicted pressure value of the corresponding area in the three-dimensional model and marks a yellow warning.

[0109] Here is a specific example:

[0110] The static load component pressure integral of a patient's sacral central subdomain was detected to be 60kPa·s, and the dynamic fluctuation component amplitude-frequency value was 8kPa·Hz (LF / HF=1.6, threshold=4×1.6=6.4kPa·Hz), triggering the "negative compensation" code. In step 2023, the weight coefficient of this subdomain is 1.0, the basic compensation amount is 60mL, the dynamic fluctuation is added by 15mL, and the total compensation amount is 75mL. Step 2024 is marked as a P1-level instruction, with a predicted deformation of 7.5mm and an effective time window of T0+10ms to T0+200ms. After the actuator was inflated with 75mL, the electronic skin feedback pressure dropped to 38kPa, and the three-dimensional model of the nurse station updated the area to a green safety state.

[0111] Through waveform decomposition, the steady-state and dynamic pressure components are precisely separated (2021), enhancing the ability to identify physiological interference; the compensation trigger mechanism based on phase synchronization and dynamic threshold (2022) significantly reduces the misjudgment rate; anatomical regional weight allocation (2023) realizes accurate spatial adaptation of pressure compensation; timing marking and deformation prediction (2024) ensure the efficiency of instruction execution and system traceability. Ultimately, the accuracy of pressure abnormality identification is increased to 95%, the matching degree of compensation operation and physiological state is improved by 60%, and the delay of model update at the nurse station is shortened to less than 500ms.

[0112] In order to solve the problems of uneven spatial response, sudden deformation boundary changes and command conflicts in multi-airbag collaborative compensation, this solution achieves smooth transition and precise coordination of airbag compensation through spatial topology modeling, attenuated gradient interpolation, timing interleaving optimization and closed-loop dynamic calibration. This design deeply integrates physical position and compensation parameters, breaking through the limitations of "single-point compensation, global interference" in traditional solutions, and forming an intelligent compensation link with both spatial continuity and timing consistency. In some embodiments, during the execution of the gradient compensation operation, the physical position coordinates of the target airbag and the difference in compensation parameters are time-interleaved to generate a trigger instruction sequence containing a three-dimensional compensation priority queue. The trigger instruction sequence forms a closed-loop link through the wireless sensor network and the multi-node data synchronization of the electronic skin, including:

[0113] 3021. Based on the physical position coordinates of the target airbag and the spatial distribution characteristics of the sacral pressure gradient trajectory, a position encoding strategy is used to generate a spatial topological relationship of the airbag group, wherein the spatial topological relationship includes radial distance parameters and axial angle parameters of adjacent airbags;

[0114] The position encoding strategy refers to generating a topological structure based on the physical coordinates of the airbags (such as the X / Y / Z values ​​in the Cartesian coordinate system). The radial distance parameter is the straight-line distance between adjacent airbags, and the axial angle parameter is the angle between the center line of the airbag group and the reference axis (such as the longitudinal axis of the human body). The spatial topological relationship is used to describe the geometric distribution characteristics of the airbag group and support the collaborative calculation of subsequent compensation parameters.

[0115] In the embodiment of the present application, the polar coordinate system is used to transform the physical coordinates of the airbags, with the sacral pressure center as the origin, the radial distance is divided into units of 10 cm, and the axial angle is divided into sectors of 30°. For example, the coordinates of a certain airbag are (radius 20 cm, angle 60°), the radial distance of its adjacent airbags is 15 cm, and the axial angle difference is 30°, forming a spatial topological relationship table.

[0116] 3022. Perform time-series interleaving processing on the compensation parameter differences, calculate the attenuation gradient of the compensation amount between adjacent airbags based on the radial distance parameter in the spatial topological relationship, generate interpolation compensation parameters for smoothing the airbag deformation boundary, and superimpose the interpolation compensation parameters with the gradient compensation parameter sequence;

[0117] The compensation parameter difference refers to the difference in target compensation between adjacent airbags in the same time window (e.g., airbag A needs to be inflated by 10%, and airbag B needs to be deflated by 5%, with a difference of 15%). The attenuation gradient calculates the transfer attenuation coefficient of the compensation amount based on the radial distance (e.g., the attenuation is 20% for every 10 cm increase in distance), and the interpolation compensation parameter is used to fill the compensation amount transition value between adjacent airbags to avoid sudden changes in deformation boundaries.

[0118] In the embodiment of the present application, for the area where the compensation difference between adjacent airbags exceeds 10%, the interpolation weight is calculated according to the radial distance. For example, airbag A (15 cm from the origin) needs to be inflated by 15%, and the adjacent airbag B (25 cm away) needs to be inflated by 5%. The radial distance difference between the two is 10 cm, and the attenuation gradient is (15%-5%)×(1-0.2×10 / 10)=8%. The interpolation compensation parameters generate three transition nodes between airbags A and B, which are inflated by 13%, 11%, and 9% respectively.

[0119] 3023. According to the superposition result of the interpolation compensation parameter and the gradient compensation parameter sequence, the compensation effective time window is phase-aligned by a time interleaving algorithm to generate a three-dimensional compensation priority queue sorted by physical position coordinates, wherein each node in the queue includes a compensation amount, an attenuation gradient, and a phase synchronization mark;

[0120] Phase alignment refers to aligning the compensation effective time windows of different airbags to the same time base after sorting them by spatial position to avoid overlapping and conflicting instructions. The three-dimensional compensation priority queue is generated according to the rule of "central area first, priority for those with large compensation difference", and the queue node contains the compensation amount, attenuation gradient value and phase synchronization mark (marking whether the instruction is synchronized with the electronic skin data).

[0121] In the embodiment of the present application, the timing interleaving algorithm divides the compensation instructions into 5ms time slices, the instructions for the central area (radius <15cm) are allocated to the first 50% of the time slices, and the instructions for the edge area are allocated to the last 50%. For example, the central airbag A (compensation amount 15%) is executed at T0-T5ms, and the edge airbag B (compensation amount 5%) is executed at T5-T10ms. The phase synchronization mark is generated by comparing the pressure change rate of the electronic skin feedback with the expected value, and the deviation <5% is marked as "synchronized".

[0122] 3024. Dynamically match the three-dimensional compensation priority queue with the multi-node pressure data collected in real time by the electronic skin, calibrate the phase synchronization mark through the wireless sensor network, and generate a trigger instruction sequence including the airbag deformation feedback parameter. The trigger instruction sequence synchronously drives the air bed actuator through a closed-loop link and updates the three-dimensional pressure model of the nurse station gateway.

[0123] Dynamic matching refers to the temporal and spatial association of the compensation parameters in the priority queue with the real-time pressure data of the electronic skin, and the timeliness of the phase synchronization mark is calibrated. The deformation feedback parameter includes the difference between the actual deformation variable and the predicted deformation variable, which is used to correct the subsequent compensation instructions. The closed-loop link realizes the continuous optimization of "execution-feedback-iteration" through real-time data reflux.

[0124] In the embodiment of the present application, when the actual deformation of a certain airbag (such as 8mm) is detected to deviate from the predicted value (7.5mm) by more than 10%, a correction instruction (such as additional inflation of 2%) is added to the trigger instruction sequence. The three-dimensional model of the nurse station updates the pressure cloud map according to the deformation feedback parameters, and the high-risk area is marked in red and a secondary compensation instruction is generated and sent to the actuator through a closed-loop link.

[0125] Here is a specific example:

[0126] Abnormal pressure gradient is detected in the central area of ​​the sacrum and coccyx of a patient. In step 3021, the spatial topological relationship between the central airbag A (radius 10cm, angle 90°) and the edge airbag B (radius 25cm, angle 90°) is generated. Step 3022 calculates the compensation difference between AB as 20%, generates interpolation parameters according to the radial distance difference of 15cm (attenuation gradient 24%), and inserts 3 transition airbags between AB to compensate 16%, 12%, and 8% respectively. Step 3023 marks the central airbag A as P1 priority (T0-T5ms execution), and the transition airbags are executed in sequence according to the distance sorting. After step 3024 is executed, the electronic skin feedbacks the deformation of the central area of ​​6mm (predicted value 7mm), triggering a correction instruction to add 1.5% inflation, and the model of the nurse station is updated to show that the pressure gradient has dropped to a safe threshold.

[0127] Through spatial topological modeling (3021), accurate spatial adaptation of airbag compensation is achieved; attenuated gradient interpolation (3022) effectively eliminates deformation boundary mutations, improving the smoothness of pressure transition by 70%; time sequence interleaving and priority queue (3023) reduce the multi-airbag response conflict rate to less than 3%; closed-loop dynamic calibration (3024) makes the compensation accuracy reach within ±2%, and the update delay of the nurse station model is compressed to 300ms. The overall solution reduces the duration of excessive pressure in high-risk areas of pressure sores by 90% and doubles the efficiency of nursing intervention.

[0128] In order to solve the problems of high false trigger rate and response hysteresis caused by the disconnection between physiological characteristics and pressure changes in traditional pressure compensation, and single compensation direction, this solution constructs a gradient compensation mechanism based on the patient's physiological state through physiological-pressure coupling analysis, dynamic threshold adjustment, regionalized direction control and closed-loop parameter update. This design deeply integrates the time-frequency characteristics of heart rate variability and pressure fluctuations, breaks through the limitations of static thresholds and one-way compensation, and realizes accurate dynamic adaptation of compensation direction and amplitude. In some embodiments, an adaptive adjustment mechanism for the pressure gradient judgment threshold is established based on the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component. When the amplitude-frequency characteristics of the dynamic fluctuation component exceed the adaptive adjustment threshold, a trigger instruction containing the gradient compensation mode code is generated. The gradient compensation mode code corresponds to the positive and negative polarity of the pressure distribution change rate, including:

[0129] 4021. Based on the phase synchronization relationship between the dynamic fluctuation component and the heart rate variability data, generate a physiological-pressure coupling strength parameter through phase correlation analysis, wherein the physiological-pressure coupling strength parameter is used to construct an adaptive adjustment mechanism for a pressure gradient determination threshold;

[0130] Phase synchronization refers to the consistency of the phase difference between the dynamic fluctuation component (such as the pressure fluctuation signal) and the heart rate variability (HRV) signal in the time and frequency domain. The physiological-pressure coupling strength parameter is obtained by calculating the cumulative variance of the phase difference between the two signals, which is used to quantify the driving strength of physiological activities on pressure fluctuations. The adaptive adjustment mechanism dynamically scales the pressure gradient judgment threshold according to the coupling strength.

[0131] In the embodiment of the present application, the Hilbert transform is used to extract the instantaneous phase of the dynamic fluctuation component and the HRV signal, and the phase difference variance in each 5-second window is calculated. If the variance is <0.1 (strong coupling), the judgment threshold is lowered by 30%; if the variance is >0.3 (weak coupling), the threshold is raised by 20%. For example, when the coupling is strong, the threshold is reduced from 8kPa·Hz to 5.6kPa·Hz, which is more sensitive to trigger compensation.

[0132] 4022. Extract high-frequency energy distribution characteristics according to the amplitude-frequency characteristics of the dynamic fluctuation component, and combine with the dynamic offset calculation model in the adaptive adjustment mechanism. When the high-frequency energy distribution characteristics exceed the adaptive adjustment threshold, generate a gradient compensation mode code bound to the positive and negative polarities of the pressure distribution change rate, and the gradient compensation mode code is used as the core parameter of the trigger instruction;

[0133] The high-frequency energy distribution feature refers to the energy integral value of the dynamic fluctuation component in the 1-2Hz frequency band, reflecting the intensity of rapid pressure changes. The dynamic offset calculation model generates a threshold offset coefficient based on the coupling strength parameter (e.g., for every 0.1 increase in coupling strength, the offset decreases by 5%). The gradient compensation mode coding defines that deflation is performed when the pressure change rate is positive (coded "+"), and inflation is performed when the rate is negative (coded "-").

[0134] In the embodiment of the present application, a short-time Fourier transform (STFT) is performed on the dynamic fluctuation component to extract the energy value in the 1-2 Hz frequency band. 2 / Hz and the coupling strength parameter is 0.8 (threshold = 6×0.8 = 4.8 kPa·Hz), triggering the compensation mode code “-” (inflation 10%). The code is also bound to the rate polarity, such as a pressure rate of -2 kPa / s corresponding to an inflation amplitude of 15%.

[0135] 4023. Performing regional mapping on the gradient compensation mode encoding according to the anatomical distribution of the multi-level support subdomains of the sacral coccyx, and combining the spatial weight distribution result of the static load component, generating compensation direction control parameters of each support subdomain, wherein the compensation direction control parameters are used to correct the action direction of the basic compensation amount;

[0136] The multi-level support subdomain refers to the independent compensation area divided by the anatomical landmarks of the sacrum and coccyx (such as the sacral crest and ischial tuberosity). The compensation direction control parameter is generated according to the polarity (+ / -) encoded by the gradient compensation mode and the spatial weight of the static load component (such as the high pressure area weight 1.5, the low pressure area 0.8), which is used to correct the direction of action of the basic compensation amount (such as the high pressure area is inflated first).

[0137] In the embodiment of the present application, the De la aunay triangulation algorithm is used to divide the sacrococcygeal part into 6 supporting subdomains. If the static load weight of subdomain A is 1.2 and the compensation mode is "-", the direction control parameter = 1.2 × (-1) = -1.44, the basic compensation amount (such as 50mL) × -1.44 = -72mL (inflated 72mL). The weight of the adjacent subdomain B is 0.8, then the compensation amount = 50 × 0.8 × (-1) = -40mL.

[0138] 4024. Based on the superposition relationship between the gradient compensation mode encoding and the compensation direction control parameter, a trigger instruction including the compensation effective direction and amplitude is generated. The trigger instruction synchronously drives the air mattress actuator to perform directional gradient compensation through the wireless sensor network closed-loop link, and is fed back to the nurse station gateway to update the dynamic response parameters of the three-dimensional pressure model.

[0139] The compensation effective direction is determined by the gradient compensation mode coding polarity ("+" for deflation, "-" for inflation), and the amplitude is calculated by the product of the compensation direction control parameter and the basic compensation amount. The dynamic response parameters include the pressure gradient change rate and HRV trend after compensation, which are used for real-time rendering and early warning of the three-dimensional model of the nurse station.

[0140] In the embodiment of the present application, the trigger instruction adopts the structured data format: {"direction":"-","amplitude":72mL,"area":"A","HRV trend":-10%}. After the actuator is inflated with 72mL, the electronic skin feedback pressure drops by 15kPa, and the LF / HF ratio of HRV drops by 0.3. The nurse station model updates area A in green, and the dynamic response parameter is marked "pressure gradient rate -1.2kPa / s".

[0141] Here is a specific example:

[0142] A dynamic fluctuation component high-frequency energy of 15 kPa was detected in the sacral crest of a patient 2 / Hz, HRV phase difference variance 0.08 (strong coupling), adaptive threshold adjusted to 4.8kPa·Hz. Step 4022 generates compensation mode code "-", binding pressure rate -2.5kPa / s. In step 4023, the static load weight of this subdomain is 1.5, the direction control parameter = -1.5, and the basic compensation amount is 60mL×-1.5 = -90mL (inflated 90mL). Step 4024 triggers the instruction to drive the actuator to inflate 90mL, the electronic skin feedback pressure drops by 18kPa, and the LF / HF of HRV drops from 1.8 to 1.5. The nurse station model marks the pressure gradient rate of this area as -1.8kPa / s and turns off the warning.

[0143] By dynamically adjusting the threshold value (4021) through physiological-pressure coupling strength, the compensation trigger sensitivity is increased by 40%; the high-frequency energy and dynamic offset model (4022) reduces the false trigger rate from 15% to 5%; the anatomical regionalization direction control (4023) achieves accurate matching of the compensation direction and body pressure distribution, and the uniformity is improved by 60%; the directional gradient compensation and closed-loop feedback (4024) reduce the duration of pressure exceeding the standard by 80%, and the update delay of the nurse station model is compressed to 200ms. The overall solution makes the accuracy of pressure sore risk area identification reach 98%, and the efficiency of nursing intervention is increased by 3 times.

[0144] In order to solve the problem of uneven pressure distribution caused by poor anatomical adaptability and rigid weight distribution in traditional pressure compensation, this solution constructs a multi-level precise compensation mechanism based on physiological structure characteristics through anatomical region segmentation, dynamic weight distribution, spatial smoothness constraint and vector superposition. This design deeply integrates static load distribution with dynamic fluctuation characteristics, breaks through the limitations of uniform weight compensation, and achieves anatomical adaptability and dynamic smoothness of pressure regulation. In some embodiments, the static load component is divided into multiple levels of support subdomains according to the anatomical region of the sacrum and coccyx, and the static load component of each support subdomain is allocated compensation weights in combination with the gradient compensation mode encoding to generate a basic compensation amount bound to the physical position of the target airbag. The basic compensation amount is superimposed with the compensation correction amount of the dynamic fluctuation component to form a gradient compensation parameter sequence, including:

[0145] 5021. Based on the physiological structure characteristics of the sacral anatomical region, a regional adaptive segmentation algorithm is used to perform multi-level support subdomain division on the static load component, and a subdomain division result including partition contour parameters and pressure distribution density is generated, wherein the partition contour parameters are bound to the physical position coordinates of the target airbag;

[0146] The regional adaptive segmentation algorithm refers to dividing the static load component into multiple independent support subdomains based on the anatomical landmarks of the sacrococcygeal bones (such as the sacral crest and ischial tuberosity) and the pressure distribution density. The partition contour parameters define the geometric boundaries of the subdomain (such as the coordinates of the polygon vertices), and the pressure distribution density is the integral value of the pressure per unit area. The spatial overlap between the physical position coordinates of the target airbag and the partition contour determines the subdomain to which it belongs.

[0147] In the embodiment of the present application, a density clustering algorithm (DBSCAN) is used to analyze the high pressure density area (>40 kPa / cm 2 ) to generate three core sub-regions: the central area of ​​the sacrum (outline is an ellipse), the left / right ischial tuberosity area (outline is a circle). Each sub-region is bound to three target airbags, and the coordinates are within 10 cm of the center of mass of the outline.

[0148] 5022. According to the positive and negative polarity characteristics encoded in the gradient compensation mode, a compensation weight dynamic allocation model of the support subdomain is established, and an initial compensation weight coefficient of each subdomain is generated in combination with the pressure distribution density. The initial compensation weight coefficient is dynamically attenuated along with the compensation correction amount of the dynamic fluctuation component.

[0149] The compensation weight dynamic allocation model adjusts the subdomain weight coefficient according to the polarity of the gradient compensation mode encoding (positive for deflation, negative for inflation). The higher the pressure distribution density, the greater the weight. Dynamic attenuation means that the weight coefficient decreases linearly with the increase of the compensation correction amount of the dynamic fluctuation component (for example, the weight decays by 0.5% for every 1% increase in the correction amount).

[0150] In the embodiment of the present application, the pressure density in the central area of ​​the sacrum is 50 kPa / cm 2 , initial weight coefficient 1.5; left ischial tuberosity density 30 kPa / cm 2 , weight 1.0. If the gradient compensation mode is negative (inflation), the weight coefficient is adjusted to 1.5×(-1)=-1.5. When the subdomain dynamic fluctuation correction amount is increased by 20%, the weight decays to -1.5×(1-0.2×20%)=-1.44.

[0151] 5023. Performing spatial continuity constraint processing on the initial compensation weight coefficient, calculating a weight smoothing factor based on partition contour parameters of adjacent support subdomains, and generating a modified dynamic compensation weight coefficient, wherein the dynamic compensation weight coefficient is used to calculate a basic compensation amount bound to a target airbag;

[0152] The spatial continuity constraint calculates the weight smoothing factor by the overlapping area of ​​the adjacent subdomain contours (for example, the smoothing factor is 0.8 when the overlapping area ratio is >30%). The dynamic compensation weight coefficient = initial weight × smoothing factor, ensuring that the difference in weights of adjacent subdomains does not exceed the preset threshold (such as ±20%).

[0153] In the embodiment of the present application, the overlap area between the sacral central area and the left ischial tuberosity area accounts for 25%, and the smoothing factor is 0.9. The dynamic compensation weight coefficient is corrected from -1.44 to -1.44×0.9=-1.296. There is no overlap in the right ischial tuberosity area, and the original weight is maintained at -1.0.

[0154] 5024. Perform vector superposition of the basic compensation amount and the compensation correction amount of the dynamic fluctuation component according to the polarity direction encoded in the gradient compensation mode to generate a gradient compensation parameter sequence including the airbag position code and the weight attenuation parameter. The gradient compensation parameter sequence drives the air bed actuator by triggering the instruction set, and synchronously updates the regional pressure distribution parameters of the three-dimensional pressure model in the nurse station gateway.

[0155] Vector superposition refers to the algebraic synthesis of the basic compensation amount and the correction amount according to the polarity direction (positive / negative) encoded in the gradient compensation mode. The weight decay parameter records the dynamic decay ratio and is used for the nurse station model to trace the history of compensation adjustment. The regional pressure distribution parameters include the pressure gradient value and density change rate after compensation.

[0156] In the embodiment of the present application, the basic compensation amount of the central area of ​​the sacrum = 100mL × |-1.296| = 129.6mL (inflation), the dynamic fluctuation correction amount is added by 25mL, and the total compensation amount is 154.6mL. The gradient compensation parameter sequence is recorded as {"area A1", "inflation 154.6mL", "attenuation -14.4%"}, and the pressure density of area A1 in the nurse station model is synchronously updated to drop to 28kPa / cm 2 .

[0157] Here is a specific example:

[0158] The static load component of the central sacrum of a patient is 55 kPa / cm 2 , in step 5021, it is divided into elliptical subdomains and airbags A1-A3 are bound. Step 5022 generates an initial weight of -1.5, and the weight decays to -1.5×0.82=-1.23 after the dynamic fluctuation correction is added by 18%. Step 5023 Due to the 20% overlap with the left ischial subdomain, the smoothing factor is 0.85, and the dynamic weight = -1.23×0.85=-1.045. In step 5024, the basic compensation amount = 100mL×1.045=104.5mL (inflation), and the total compensation amount is 124.5mL after the correction amount of 20mL is superimposed. After the actuator is inflated, the electronic skin feedback pressure density drops to 30kPa / cm 2 , the nurse station model marks the pressure gradient rate in this area as -1.5 kPa / s.

[0159] Through anatomical region segmentation and contour binding (5021), the matching degree between the compensation region and the physiological structure is improved by 90%; dynamic weight allocation and attenuation mechanism (5022-5023) realizes precise anatomical adaptation of pressure regulation, and the compensation uniformity is improved by 65%; vector superposition and parameter tracing (5024) make the pressure distribution update accuracy of the nurse station model reach ±3kPa, and the closed-loop response time is shortened to 150ms. The overall solution reduces the duration of excessive pressure in high-risk areas of pressure sores by 85%, and improves the accuracy of nursing intervention by 4 times.

[0160] In order to solve the problem of model update lag and execution asynchrony caused by data matching mismatch and instruction delay in traditional pressure compensation, this solution constructs a real-time collaborative control link of airbag deformation and pressure distribution through dynamic data matching, phase delay calibration, closed-loop incremental optimization and multi-source data fusion. This design deeply couples the priority queue with real-time sensor data, breaks through the limitations of one-way instruction execution, and realizes the full closed-loop precision control of "perception-decision-execution-verification". In some embodiments, the three-dimensional compensation priority queue is dynamically matched with the multi-node pressure data collected in real time by the electronic skin, and the phase synchronization mark is calibrated through the wireless sensor network to generate a trigger instruction sequence containing airbag deformation feedback parameters. The trigger instruction sequence synchronously drives the air mattress actuator through a closed-loop link and updates the three-dimensional pressure model of the nurse station gateway, including:

[0161] 6021. Based on the compensation amount, attenuation gradient and phase synchronization marking parameters in the three-dimensional compensation priority queue, a dynamic matching rule with the multi-node pressure data of the electronic skin is established, and a mapping relationship between the queue node and the pressure collection point is generated by measuring the similarity of the pressure distribution. The mapping relationship is used to extract the initial calibration value of the airbag deformation feedback parameter;

[0162] The dynamic matching rule is to establish a mapping relationship between queue nodes and pressure collection points by calculating the spatial distribution similarity of compensation parameters (compensation amount, attenuation gradient) in the three-dimensional compensation priority queue and electronic skin pressure data (such as pressure value distribution form, gradient direction consistency). The initial calibration value is the initial deviation between the predicted value of the airbag shape variable extracted according to the mapping relationship and the actual value.

[0163] In the embodiment of the present application, a pressure distribution morphology matching algorithm is used to compare the similarity between the compensation amount distribution of the queue node (such as 15% compensation amount in the center area and 5% at the edge) and the real-time pressure distribution of the electronic skin (45kPa in the center and 20kPa at the edge). If the similarity is >85%, the mapping relationship is marked as valid, and the initial calibration value = (predicted deformation 7mm-measured deformation 6.5mm) = 0.5mm.

[0164] 6022. According to the timing offset between the phase synchronization mark parameter and the real-time pressure data of the electronic skin, a phase compensation algorithm is used to dynamically calibrate the effective time window of the trigger instruction sequence, and a corrected trigger instruction sequence including the calibrated phase delay parameter is generated, wherein the corrected trigger instruction sequence matches the response delay characteristic of the air mattress actuator;

[0165] The timing offset refers to the difference between the expected effective timestamp of the trigger instruction and the actual effective timestamp fed back by the electronic skin (e.g. expected execution at T0+50ms, actual feedback at T0+60ms, offset 10ms). The phase compensation algorithm dynamically adjusts the effective time window of subsequent instructions according to the offset (e.g., send it 10ms in advance), and the calibrated phase delay parameter records the time correction after compensation.

[0166] In the embodiment of the present application, when the timing offset is detected to be 10ms continuously, the phase compensation algorithm will uniformly advance the subsequent instruction effective time window by 10ms. A new field "Phase Delay -10ms" is added to the modified trigger instruction sequence to ensure that the actuator starts to act at T0+40ms, which is strictly synchronized with the electronic skin feedback data.

[0167] 6023. Combining the calibrated phase delay parameter and the airbag deformation feedback parameter in the correction trigger instruction sequence, dynamically adjusting the attenuation gradient and the compensation amount through the feedback mechanism of the closed-loop link, generating an optimized trigger instruction sequence including the deformation compensation increment parameter, and the optimized trigger instruction sequence synchronously updates the node parameters of the three-dimensional compensation priority queue;

[0168] The deformation compensation increment parameter refers to the compensation correction value (such as additional inflation of 1%) dynamically calculated according to the deformation feedback deviation (such as predicted deformation 8mm vs. measured 7.2mm). The attenuation gradient and compensation amount are adjusted through the proportional-integral (PI) control algorithm, and the trigger instruction sequence is optimized to synchronously update the compensation amount and phase mark of the three-dimensional queue node.

[0169] In the embodiment of the present application, if the deformation feedback deviation is 0.8mm (target deformation 8mm), the system generates incremental parameters according to the deviation ratio: compensation amount is added (0.8 / 8)×100%=10%. The attenuation gradient is adjusted from 20% to 18% (reduced by 2% for smooth transition), and the optimization instruction sequence updates the queue node parameters to "compensation amount +10%, attenuation gradient 18%".

[0170] 6024. Data fusion is performed between the optimized trigger instruction sequence and the three-dimensional pressure model of the nurse station gateway, and the regional pressure distribution value of the pressure model is dynamically corrected based on the deformation compensation incremental parameter. The corrected pressure distribution value is fed back to the electronic skin network through a closed-loop link to form real-time coordinated control of airbag deformation and pressure distribution.

[0171] Data fusion refers to weighted superposition of the compensation increment parameters in the optimization trigger instruction and the pressure prediction value of the three-dimensional pressure model (such as 70% weight of the incremental parameter and 30% weight of the historical data). The dynamically corrected pressure distribution value is fed back to the electronic skin network through a closed-loop link, triggering the generation of the next round of compensation instructions.

[0172] In the embodiment of the present application, after receiving the optimization instruction, the three-dimensional model of the nurse station corrects the predicted pressure value of area A from 50kPa to 50×(1-10%)=45kPa. The correction value is fed back to the electronic skin through the wireless network. If the real-time pressure detection value is 44kPa (error 1kPa), a secondary compensation instruction is generated to add 2% of inflation.

[0173] Here is a specific example:

[0174] The preset compensation amount of the three-dimensional queue node in the sacral center area of ​​a patient is 20% (predicted deformation amount 10mm), and the electronic skin detects a deformation amount of 9mm in real time (initial calibration value in step 6021 + 1mm). Step 6022 detects the timing offset of 15ms, and the instruction is issued 15ms in advance after calibration. Step 6023 generates incremental parameters to add compensation amount of 2% (total compensation amount 22%), and the deformation prediction value is corrected to 10.2mm. In step 6024, the nurse station model corrects the regional pressure value from 55kPa to 53.8kPa, and the electronic skin detects the value of 53.5kPa in the next cycle, triggering a secondary compensation of 0.5% to form a closed-loop control.

[0175] Through dynamic matching and phase calibration (6021-6022), the synchronization accuracy of command and actuator response is improved to ±3ms; closed-loop incremental optimization (6023) compresses the deformation prediction error from 12% to less than 3%; data fusion and real-time feedback (6024) make the pressure distribution update delay of the nurse station model less than 200ms, and the pressure control response speed is increased by 4 times. Ultimately, the millimeter-level precise coordinated control of airbag deformation and pressure distribution is achieved, and the accuracy rate of pressure sore risk area identification reaches 99%.

[0176] Figure 2 A schematic diagram of a wireless network-based nurse station ward information collection system is provided for an embodiment of the present application. Figure 2 As shown, the device comprises:

[0177] A capture module 21 is used to integrate a flexible electronic skin network inside the mattress, capture the sacral pressure distribution through a distributed piezoresistive unit, and fuse the impedance and temperature parameters to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol;

[0178] A driving module 22, for generating a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, driving the air mattress actuator to perform a gradient compensation operation on the target airbag, wherein the trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network;

[0179] The allocation module 23 is used to perform spatiotemporal encoding of the airbag deformation parameters and the body position deviation angle in the compensation operation, generate hierarchical data packets through a compression algorithm, and transmit them to the nurse station gateway via a remote wide area network protocol using dynamic channel allocation. The nurse station gateway is configured with a multi-band aggregation module to adapt to the hospital wireless network;

[0180] The reconstruction module 24 is used to parse the spatiotemporal coding in the hierarchical data packet at the nurse station terminal, reconstruct the three-dimensional pressure model of the airbag deformation and body position displacement based on the wireless transmission quality, generate early warning instructions containing the pressure change rate and intervention duration in combination with the vital sign file, and map it into a dynamic thermal topology map through the wireless broadcast mechanism.

[0181] Figure 2 The wireless network-based nurse station ward information collection device can perform Figure 1 The implementation principle and technical effect of the wireless network-based nurse station ward information collection method described in the illustrated embodiment will not be repeated. The specific manner in which each module and unit performs operations in the wireless network-based nurse station ward information collection device in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0182] In one possible design, Figure 2 The wireless network-based nurse station ward information collection device of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0183] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0184] The processing component 32 is used for the above Figure 1 The embodiment provides a method for collecting ward information at a nurse station based on a wireless network.

[0185] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0186] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0187] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0188] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0189] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0190] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0191] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a method for collecting ward information at a nurse station based on a wireless network.

[0192] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0193] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0194] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for collecting ward information of a nurse station based on a wireless network, characterized in that: The following steps are involved: A flexible electronic skin network is integrated inside the mattress, and the pressure distribution of the sacrum and coccyx is captured by a distributed piezoresistive unit. The impedance and temperature parameters are fused to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol. Generate a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, drive the air mattress actuator to perform gradient compensation operation on the target airbag, and the trigger instruction forms a closed-loop link with the electronic skin through the wireless sensor network; The airbag deformation parameters and body position deviation angles in the compensation operation are encoded in time and space, and hierarchical data packets are generated through compression algorithms. Dynamic channel allocation is used to transmit the data packets to the nurse station gateway via the remote wide area network protocol. The nurse station gateway is equipped with a multi-band aggregation module to adapt to the hospital wireless network. The spatiotemporal coding in the hierarchical data packet is parsed at the nurse station terminal, and the three-dimensional pressure model of the airbag deformation and body position displacement is reconstructed based on the wireless transmission quality. The early warning instruction containing the pressure change rate and intervention duration is generated in combination with the vital sign file, and is mapped into a dynamic thermal topology map through the wireless broadcast mechanism.

2. The method according to claim 1, characterized in that Based on the pressure gradient trajectory in the multimodal thermal map and the matching of physiological parameters, a trigger instruction is generated to drive the air mattress actuator to perform a gradient compensation operation on the target airbag, including: Based on the periodic morphological classification results of the sacral and coccygeal pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into a static load component and a dynamic fluctuation component, and a trigger instruction is generated by dynamically adjusting the pressure gradient determination threshold; According to the airbag basic compensation amount and compensation correction amount generated by the static load component and the dynamic fluctuation component, a dynamic mapping relationship between the target airbag pressure compensation parameter and the body support surface is established, and the air mattress actuator is driven to perform the gradient compensation operation through the gradient compensation parameter sequence; During the execution of the gradient compensation operation, the physical position coordinates of the target airbag and the compensation parameter difference are processed in time sequence to generate a trigger instruction sequence including a three-dimensional compensation priority queue; The gradient compensation parameter sequence and the trigger instruction sequence are dynamically coupled to form a composite instruction set, which synchronously includes the airbag deformation prediction parameters and the physiological response dynamic parameters, which are transmitted to the air bed actuator through the closed-loop link to perform the gradient compensation operation, and fed back to the nurse station gateway to reconstruct the pressure model.

3. The method according to claim 2, characterized in that Based on the periodic morphological classification results of the sacral and coccygeal pressure gradient trajectory in the multimodal thermal map and the heart rate variability data in the physiological parameters, the continuous pressure waveform is decomposed into a static load component and a dynamic fluctuation component, and a trigger instruction is generated by dynamically adjusting the pressure gradient judgment threshold, including: Based on the periodic morphological classification results of the sacral pressure gradient trajectory, a waveform morphological matching algorithm is used to extract low-frequency baselines and decompose high-frequency disturbances of the continuous pressure waveform to generate static load components and dynamic fluctuation components; According to the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component, an adaptive adjustment mechanism for the pressure gradient determination threshold is established, and when the amplitude-frequency characteristic of the dynamic fluctuation component exceeds the adaptive adjustment threshold, a trigger instruction containing the gradient compensation mode code is generated.

4. The method according to claim 2, characterized in that: The physical position coordinates of the target airbag and the compensation parameter difference are processed in time sequence to generate a trigger instruction sequence containing a three-dimensional compensation priority queue, including: Based on the physical position coordinates of the target airbag and the spatial distribution characteristics of the sacral pressure gradient trajectory, a position encoding strategy is used to generate a spatial topological relationship of the airbag group, wherein the spatial topological relationship includes radial distance parameters and axial angle parameters of adjacent airbags; Performing time-series interleaving processing on the compensation parameter differences, calculating the attenuation gradient of the compensation amount between adjacent airbags based on the radial distance parameter in the spatial topological relationship, generating interpolation compensation parameters for smoothing the airbag deformation boundary, and superimposing the interpolation compensation parameters with the gradient compensation parameter sequence; According to the superposition result of the interpolation compensation parameter and the gradient compensation parameter sequence, the compensation effective time window is phase aligned by a time interleaving algorithm to generate a three-dimensional compensation priority queue sorted by physical position coordinates, wherein each node in the queue contains a compensation amount, an attenuation gradient and a phase synchronization mark; The three-dimensional compensation priority queue is dynamically matched with the multi-node pressure data collected by the electronic skin in real time, and the phase synchronization mark is calibrated through the wireless sensor network to generate a trigger instruction sequence containing the airbag deformation feedback parameters.

5. The method according to claim 3, characterized in that: According to the phase synchronization relationship between the heart rate variability data and the dynamic fluctuation component, an adaptive adjustment mechanism for the pressure gradient determination threshold is established. When the amplitude-frequency characteristic of the dynamic fluctuation component exceeds the adaptive adjustment threshold, a trigger instruction containing the gradient compensation mode code is generated, including: Based on the phase synchronization relationship between the dynamic fluctuation component and the heart rate variability data, a physiological-pressure coupling strength parameter is generated through phase correlation analysis, and the physiological-pressure coupling strength parameter is used to construct an adaptive adjustment mechanism for the pressure gradient determination threshold; The high-frequency energy distribution feature is extracted according to the amplitude-frequency characteristics of the dynamic fluctuation component, and combined with the dynamic offset calculation model in the adaptive adjustment mechanism, when the high-frequency energy distribution feature exceeds the adaptive adjustment threshold, a gradient compensation mode code bound to the positive and negative polarities of the pressure distribution change rate is generated; The gradient compensation mode encoding is regionally mapped according to the anatomical distribution of the multi-level support subdomains of the sacral coccyx, and the compensation direction control parameters of each support subdomain are generated in combination with the spatial weight distribution result of the static load component; Based on the superposition relationship between the gradient compensation mode code and the compensation direction control parameter, a trigger instruction including the compensation effective direction and amplitude is generated.

6. The method according to claim 3, characterized in that: Also includes: Based on the physiological structural characteristics of the sacral anatomical region, a regional adaptive segmentation algorithm is used to divide the static load component into multi-level support subdomains, generating subdomain division results including partition contour parameters and pressure distribution density. According to the positive and negative polarity characteristics encoded by the gradient compensation mode, a compensation weight dynamic allocation model of the support subdomain is established, and the initial compensation weight coefficient of each subdomain is generated in combination with the pressure distribution density; Performing spatial continuity constraint processing on the initial compensation weight coefficient, calculating a weight smoothing factor based on partition contour parameters of adjacent supporting subdomains, and generating a revised dynamic compensation weight coefficient; The basic compensation amount and the compensation correction amount of the dynamic fluctuation component are vector-superimposed in the polarity direction of the gradient compensation mode encoding to generate a gradient compensation parameter sequence including the airbag position encoding and the weight attenuation parameter.

7. The method according to claim 4, characterized in that The three-dimensional compensation priority queue is dynamically matched with the multi-node pressure data collected by the electronic skin in real time, and the phase synchronization mark is calibrated through the wireless sensor network to generate a trigger instruction sequence containing the airbag deformation feedback parameter, including: Based on the compensation amount, attenuation gradient and phase synchronization marking parameters in the three-dimensional compensation priority queue, a dynamic matching rule with the multi-node pressure data of the electronic skin is established, and a mapping relationship between the queue node and the pressure collection point is generated by measuring the similarity of the pressure distribution; According to the timing offset between the phase synchronization mark parameter and the real-time pressure data of the electronic skin, a phase compensation algorithm is used to dynamically calibrate the effective time window of the trigger instruction sequence to generate a modified trigger instruction sequence including the calibrated phase delay parameter; Combined with the calibrated phase delay parameter and the airbag deformation feedback parameter in the modified trigger instruction sequence, the attenuation gradient and the compensation amount are dynamically adjusted through the feedback mechanism of the closed-loop link to generate an optimized trigger instruction sequence including the deformation compensation increment parameter.

8. A wireless network-based nurse station ward information collection system, characterized in that: include: A capture module is used to integrate a flexible electronic skin network inside a mattress, capture the sacral pressure distribution through a distributed piezoresistive unit, and fuse the impedance and temperature parameters to generate a multi-modal thermal map. The flexible electronic skin network realizes multi-node data synchronization through a wireless ad hoc network protocol; A driving module, for generating a trigger instruction based on the pressure gradient trajectory in the multimodal thermal map and matching the physiological parameters, driving the air mattress actuator to perform a gradient compensation operation on the target airbag, wherein the trigger instruction forms a closed-loop link with the electronic skin through a wireless sensor network; The allocation module is used to perform spatiotemporal encoding on the airbag deformation parameters and the body position deviation angle in the compensation operation, generate hierarchical data packets through a compression algorithm, and transmit them to the nurse station gateway via a remote wide area network protocol using dynamic channel allocation. The nurse station gateway is configured with a multi-band aggregation module to adapt to the hospital wireless network; The reconstruction module is used to parse the spatiotemporal coding in the hierarchical data packet at the nurse station terminal, reconstruct the three-dimensional pressure model of the airbag deformation and body position displacement based on the wireless transmission quality, generate early warning instructions containing the pressure change rate and intervention duration in combination with the vital sign file, and map it into a dynamic thermal topology map through the wireless broadcast mechanism.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a wireless network-based nurse station ward information collection method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a method for collecting ward information of a nurse station based on a wireless network as described in any one of claims 1 to 7 is implemented.

Images