Intelligent rehabilitation bed body dynamic supporting system based on pressure feedback

Abstract

The invention discloses an intelligent rehabilitation bed body dynamic supporting system based on pressure feedback, and relates to the technical field of intelligent rehabilitation bed body dynamic supporting, and the intelligent rehabilitation bed body dynamic supporting system comprises a sensing acquisition module, a data fusion module, an individual modeling module, a strategy generation module, an execution driving module, a state feedback module and a remote interaction module; the sensing and collecting module is used for collecting pressure, temperature, humidity and heat distribution information of a contact area of a bedridden patient and a bed body in real time in a distributed multi-mode sensing mode and outputting original sensing signals. The data fusion module is used for performing multi-source data fusion processing on the original sensing signals from the sensing acquisition module by adopting a Kalman filtering algorithm, extracting comprehensive health indexes reflecting the microcirculation state of local tissues, and outputting fused physiological evaluation data; and the individual modeling module is used for training the body type, the weight, the medical history information and the historical pressure distribution data of the patient by adopting a machine learning model.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic support for intelligent rehabilitation beds, in particular to a dynamic support system for intelligent rehabilitation beds based on pressure feedback. Background Art

[0002] Intelligent rehabilitation bed dynamic support technology is a new type of rehabilitation assistive device that integrates multiple advanced technologies. Therefore, how to use advanced technologies to improve the intelligence level and safety of intelligent rehabilitation bed dynamic support has become one of the most pressing issues to be addressed.

[0003] In the field of dynamic support for intelligent rehabilitation beds, traditional nursing methods often rely on manual adjustment of the patient's position, which makes it difficult to achieve continuous monitoring and timely adjustment, resulting in long-term bedridden patients being prone to pressure sores. In addition, different patients have large differences in body shape, weight and health status, but most existing rehabilitation beds use a unified standard for support adjustment, which cannot meet individual needs. At the same time, traditional mechanical adjustment devices are slow to respond and cannot quickly adjust according to the patient's immediate physiological changes, affecting the treatment effect. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent rehabilitation bed dynamic support system based on pressure feedback to solve the problem that traditional nursing methods often rely on manual adjustment of patient position, which makes it difficult to achieve continuous monitoring and timely adjustment, resulting in long-term bedridden patients being prone to pressure sores, and different patients have large differences in body shape, weight and health status. However, most existing rehabilitation beds use a unified standard for support adjustment, which cannot meet individual needs. At the same time, traditional mechanical adjustment devices are slow to respond and cannot quickly adjust according to the patient's immediate physiological changes, affecting the treatment effect.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides an intelligent rehabilitation bed dynamic support system based on pressure feedback, which comprises: Perception and acquisition module, data fusion module, individual modeling module, strategy generation module, execution drive module, state feedback module, and remote interaction module; The sensing and acquisition module is used to collect the pressure, temperature, humidity and heat distribution information of the contact area between the bedridden patient and the bed in real time using a distributed multimodal sensing method, and output the original sensing signal; The data fusion module is used to perform multi-source data fusion processing on the original sensor signals from the sensing acquisition module using a Kalman filter algorithm, extract comprehensive health indicators reflecting the microcirculation status of local tissues, and output fused physiological assessment data; The individual modeling module is used to use a machine learning model to train the patient's body shape, weight, medical history information and historical pressure distribution data to establish a personalized pressure-sensitive area model and output personalized support feature parameters; The strategy generation module is used to use a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module and the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure sores, generate a dynamic support adjustment strategy, and output a support action instruction sequence; The execution drive module is used to receive the support action instruction sequence output by the strategy generation module using a partitioned independent drive unit composed of a flexible pneumatic actuator and a shape memory alloy, and implement millimeter-level deformation control on the corresponding area of ​​the bed to achieve dynamic support adjustment on demand; The state feedback module is used to resample the contact state between the bed and the human body after the adjustment is completed by the execution drive module by calling the perception acquisition module again, and compare and analyze the data before and after the adjustment to output the adjustment effect evaluation result; The remote interaction module is used to upload the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to a remote terminal device for medical staff to view and set personalized intervention parameters.

[0007] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the sensing acquisition module uses a distributed multimodal sensing method to collect pressure, temperature, humidity and heat distribution information of the contact area between the bedridden patient and the bed in real time, and outputs the original sensing signal, including: Multiple sensor nodes are arranged on the surface of the bed, each node integrates a pressure sensor, an infrared thermopile sensor, a humidity sensor and a temperature sensor; Each sensor node converts the collected physical quantity into an electrical signal and converts it into a digital signal through an A / D converter; The pressure sensor is used to measure the pressure value at each point of the contact surface between the human body and the bed; Infrared thermopile sensor is used to measure the thermal radiation intensity of the local skin surface , and converted to equivalent skin temperature; Humidity sensor is used to measure the relative humidity of the bed surface microenvironment ; Temperature sensor used to measure ambient temperature ; All collected data are packaged as raw sensor signals , and transmitted to the data fusion module for subsequent processing.

[0008] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the data fusion module uses the Kalman filter algorithm to perform multi-source data fusion processing on the original sensor signals from the sensing acquisition module, extracts comprehensive health indicators reflecting the local tissue microcirculation status, and outputs the fused physiological assessment data, including: Construct the state equation, the expression is: ; in, is the state vector, which contains the pressure, temperature, humidity and ambient temperature estimates at the current moment, is the state transition matrix, is the control input matrix, is the external disturbance input, is the process noise, which follows a normal distribution ; The observation equation is: ; in, is the observation vector, i.e., the actual sensor reading, is the observation matrix, is the observation noise, which follows a normal distribution ; The Kalman gain KkKk is calculated as follows: ; The updated state estimate is: ; Output fused physiological assessment data , which represents the pressure, temperature, humidity and ambient temperature values ​​at each point after filtering optimization.

[0009] As a preferred embodiment of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the individual modeling module uses a machine learning model to train the patient's body shape, weight, medical history information and historical pressure distribution data to establish a personalized pressure sensitive area model and output personalized support characteristic parameters, including: Input includes basic patient information, including weight, height, age, and medical history; Simultaneously input historical pressure distribution data , indicating the past The pressure value of each sensor at a time point; Use convolutional neural network (CNN) to extract features from the spatial pressure distribution map and generate spatial feature vectors; Use the long short-term memory network LSTM to model time series data and generate time feature vectors; The spatial feature vector and the temporal feature vector are concatenated and input into the fully connected layer to output the personalized support feature parameters. ,in Indicates the The pressure sensitivity weight of each area; Output It is sent to the strategy generation module to guide subsequent dynamic support adjustments.

[0010] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the strategy generation module uses a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module and the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure sores, generate a dynamic support adjustment strategy, and output a support action instruction sequence, including: The reinforcement learning model uses the deep Q network DQN to define the state space Pressure distribution diagram after fusion ; The action space is defined as the direction and amplitude of deformation adjustment of each partition of the bed; The reward function is designed as follows: ; in, is the individualized support feature parameter, is the current pressure value, is the target pressure range; In each decision cycle, the current state is input and support feature parameters , output the optimal action sequence ; The supporting action instruction sequence is encoded into an instruction format that can be recognized by the execution driver module.

[0011] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the execution drive module uses a partitioned independent drive unit composed of a flexible pneumatic actuator and a shape memory alloy to receive the support action instruction sequence output by the strategy generation module, and implements millimeter-level deformation control on the corresponding area of ​​the bed to achieve on-demand dynamic support adjustment, including: The bed is divided into Independent control areas, each area is equipped with a flexible pneumatic actuator and a shape memory alloy wire; The pneumatic actuator is supplied with air by a micro air pump, and the air intake / exhaust volume is controlled by a solenoid valve, thereby changing the height of the area; The shape memory alloy wire is controlled by a constant current power supply, and after being energized, a phase change occurs to generate a contraction force, thereby driving the local bed plate to deform; The control logic is as follows: ; in, It is Action instructions for the region, It is the displacement increment set according to the action type; The flexible material ensures smooth transition of deformation to avoid abrupt changes.

[0012] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback of the present invention, the state feedback module re-samples the contact state between the bed and the human body after the execution drive module completes the adjustment by calling the perception acquisition module again, and compares and analyzes the data before and after the adjustment, and outputs the adjustment effect evaluation result, including: After the adjustment is completed, the sensing acquisition module is started again to obtain new original sensor signals; Compare point by point with the data before adjustment and calculate the pressure change rate: ; Compare the temperature changes at each point , humidity changes ; The comprehensive evaluation formula is as follows: ; in, , , is the weighting coefficient, It is a comprehensive score, the smaller it is, the better the adjustment effect is; Output the adjustment effect evaluation results for the remote interaction module to upload to the medical staff terminal.

[0013] As a preferred solution of the intelligent rehabilitation bed dynamic support system based on pressure feedback described in the present invention, the remote interaction module uploads the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to the remote terminal device for medical staff to view and set personalized intervention parameters, including: local communication unit and cloud service platform; The local communication unit is connected to the bed control system using either Wi-Fi or Bluetooth protocols to receive the following three types of data: Physiological assessment data output by the data fusion module; The supporting action instruction sequence output by the strategy generation module; The regulation effect evaluation result output by the state feedback module; The local communication unit uploads the above data to the cloud service platform via one of the MQTT or HTTP protocols; The cloud service platform categorizes and stores data and provides a visual interface to display the patient's real-time pressure distribution thermodynamic map, temperature distribution map, humidity change curve, and ambient temperature and humidity trends; Medical staff can access the cloud platform through remote terminal devices to view historical data and set personalized intervention parameters, namely the target pressure value for each area; The personalized intervention parameters are sent back to the strategy generation module as the target basis for dynamic support adjustment in the next cycle; The remote interaction module also supports abnormal warning push function. When the pressure value of a certain area exceeds the safety threshold or the regulation effect score continues to deteriorate, an alarm notification will be automatically sent to the designated terminal device. All remote operations are subject to identity authentication and data encryption to ensure patient privacy and system security.

[0014] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the intelligent rehabilitation bed dynamic support system based on pressure feedback as described in the first aspect of the present invention is implemented.

[0015] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the intelligent rehabilitation bed dynamic support system based on pressure feedback as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are: by using the Kalman filter algorithm to process the original sensor signal, constructing the state equation and observation equation, and calculating the Kalman gain to update the state estimate, a comprehensive health indicator reflecting the microcirculation status of the local tissue is obtained. The method effectively filters the noise, improves the accuracy and reliability of the data, and enables the system to more accurately identify the health status of the local tissue. A machine learning model is used to combine the patient's body shape, weight, medical history information and historical pressure distribution data to establish a personalized pressure-sensitive area model. The method can generate support feature parameters customized according to the specific situation of each patient, ensuring a high degree of personalization of the support adjustment strategy. Using the reinforcement learning algorithm, based on the results of the data fusion module and the individual modeling module, future high-risk areas for pressure ulcers are predicted and dynamic support adjustment strategies are formulated. The method allows the system to autonomously learn and optimize its adjustment strategy to adapt to the changing needs of different patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of the intelligent rehabilitation bed dynamic support system based on pressure feedback in Example 1. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0022] Example 1, reference Figure 1 , is an embodiment of the present invention, which provides an intelligent rehabilitation bed dynamic support system based on pressure feedback, including: Perception and acquisition module, data fusion module, individual modeling module, strategy generation module, execution drive module, state feedback module, and remote interaction module; The sensing and acquisition module is used to collect the pressure, temperature, humidity and heat distribution information of the contact area between the bedridden patient and the bed in real time using a distributed multimodal sensing method, and output the original sensor signal; Furthermore, the sensing and acquisition module uses distributed multimodal sensing to collect real-time information on pressure, temperature, humidity, and heat distribution in the contact area between the bedridden patient and the bed, and outputs raw sensor signals, including: Multiple sensor nodes are arranged on the surface of the bed, each node integrates a pressure sensor, an infrared thermopile sensor, a humidity sensor and a temperature sensor; Each sensor node converts the collected physical quantity into an electrical signal and converts it into a digital signal through an A / D converter; The pressure sensor is used to measure the pressure value at each point of the contact surface between the human body and the bed; Infrared thermopile sensor is used to measure the thermal radiation intensity of the local skin surface , and converted to equivalent skin temperature; Humidity sensor is used to measure the relative humidity of the bed surface microenvironment ; Temperature sensor used to measure ambient temperature ; All collected data are packaged as raw sensor signals , and transmitted to the data fusion module for subsequent processing; It should be noted that by dynamically adjusting the distribution density of sensor nodes according to the patient's body shape and clinical needs, and increasing the number of sensors at bony protrusions, high-precision monitoring of pressure-sensitive areas is achieved. At the same time, flexible capacitive or piezoresistive pressure sensors and non-contact infrared thermopile sensors are used to ensure stability and comfort during data acquisition. The design not only improves the adaptability and measurement accuracy of the sensing system, but also avoids the compression discomfort or false alarm problems caused by traditional rigid sensors, thereby improving the patient's user experience and ultimately achieving the purpose of improving the overall perception ability of the system and enhancing the reliability of support and adjustment decisions.

[0023] The data fusion module is used to perform multi-source data fusion processing on the original sensor signals from the sensing acquisition module using the Kalman filter algorithm, extract comprehensive health indicators reflecting the microcirculation status of local tissues, and output fused physiological assessment data; Furthermore, the data fusion module uses the Kalman filter algorithm to perform multi-source data fusion processing on the original sensor signals from the perception acquisition module, extracting comprehensive health indicators reflecting the microcirculatory status of local tissues, and outputting fused physiological assessment data, including: Construct the state equation, the expression is: ; in, is the state vector, which contains the pressure, temperature, humidity and ambient temperature estimates at the current moment, is the state transition matrix, is the control input matrix, is the external disturbance input, is the process noise, which follows a normal distribution ; The observation equation is: ; in, is the observation vector, i.e., the actual sensor reading, is the observation matrix, is the observation noise, which follows a normal distribution ; The Kalman gain KkKk is calculated as follows: ; The updated state estimate is: ; Output fused physiological assessment data , represents the pressure, temperature, humidity and ambient temperature values ​​at each point after filtering optimization; It should be noted that by using Kalman filtering and its improved algorithm to fuse multi-source heterogeneous sensor signals and introducing an adaptive historical window mechanism, a high-precision assessment of the microcirculation state of local tissues is achieved. This method effectively suppresses sensor noise interference and improves the system's response sensitivity to changes in physiological parameters. The process not only improves the reliability of the data, but also provides high-quality input for subsequent individual modeling and strategy generation, ensuring the scientific nature and personalization of the support regulation strategy, and ultimately achieving the goal of improving the intelligence level of the system and enhancing the effect of rehabilitation intervention.

[0024] The individual modeling module uses a machine learning model to train the patient's body shape, weight, medical history, and historical pressure distribution data to establish a personalized pressure-sensitive area model and output personalized support feature parameters; Furthermore, the individual modeling module uses a machine learning model to train the patient's body shape, weight, medical history information, and historical pressure distribution data to establish a personalized pressure-sensitive area model and output personalized support feature parameters, including: Input includes basic patient information, including weight, height, age, and medical history; Simultaneously input historical pressure distribution data , indicating the past The pressure value of each sensor at a time point; Use convolutional neural network (CNN) to extract features from the spatial pressure distribution map and generate spatial feature vectors; Use the long short-term memory network LSTM to model time series data and generate time feature vectors; The spatial feature vector and the temporal feature vector are concatenated and input into the fully connected layer to output the personalized support feature parameters. ,in Indicates the The pressure sensitivity weight of each area; Output It is sent to the strategy generation module to guide subsequent dynamic support adjustments; It should be noted that by constructing machine learning models through local training, cloud training or federated learning, and combining CNN and LSTM to extract spatial and temporal features respectively, deep modeling of patients' body shape, weight, medical history and pressure distribution patterns is achieved. After introducing the attention mechanism, the model can automatically focus on key areas, significantly improving the ability to identify high-risk points for pressure ulcers. The personalized modeling method enables the system to customize support strategies according to the physiological characteristics of different patients, greatly improving the effectiveness and pertinence of intervention, and ultimately achieving the effect of improving nursing quality and reducing the incidence of pressure ulcers, reflecting the transformation of intelligent rehabilitation systems from "universal" to "personalized".

[0025] A strategy generation module uses a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module with the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure ulcers, generate dynamic support adjustment strategies, and output support action instruction sequences; Furthermore, the strategy generation module uses a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module with the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure ulcers, generate dynamic support adjustment strategies, and output a sequence of support action instructions, including: The reinforcement learning model uses the deep Q network DQN to define the state space Pressure distribution diagram after fusion ; The action space is defined as the direction and amplitude of deformation adjustment of each partition of the bed; The reward function is designed as follows: ; in, is the individualized support feature parameter, is the current pressure value, is the target pressure range; In each decision cycle, the current state is input and support feature parameters , output the optimal action sequence ; The supporting action instruction sequence is encoded into an instruction format that can be recognized by the execution driver module; It should be noted that by designing a deep Q-network framework based on reinforcement learning and introducing personalized support feature parameters as weight factors, the prediction of future pressure ulcer risk areas and the generation of optimal support strategies are achieved. In addition, the application of multi-agent collaborative mechanism enables the system to have knowledge sharing capabilities and accelerates the model convergence speed. The intelligent decision-making mechanism not only reduces the need for manual intervention, but also can dynamically adjust the support action according to real-time physiological data to ensure that it is always in the optimal support state, ultimately achieving the purpose of improving the system's autonomous decision-making ability and enhancing nursing efficiency and safety.

[0026] The execution drive module is used to receive the support action instruction sequence output by the strategy generation module through a partitioned independent drive unit composed of flexible pneumatic actuators and shape memory alloys, and implement millimeter-level deformation control on the corresponding area of ​​the bed to achieve dynamic support adjustment on demand; Furthermore, the execution drive module uses a partitioned independent drive unit composed of flexible pneumatic actuators and shape memory alloys to receive the support action instruction sequence output by the strategy generation module, and implement millimeter-level deformation control on the corresponding area of ​​the bed to achieve on-demand dynamic support adjustment, including: The bed is divided into Independent control areas, each area is equipped with a flexible pneumatic actuator and a shape memory alloy wire; The pneumatic actuator is supplied with air by a micro air pump, and the air intake / exhaust volume is controlled by a solenoid valve, thereby changing the height of the area; The shape memory alloy wire is controlled by a constant current power supply. When powered on, it undergoes a phase change and generates a contraction force, which drives the local bed plate to deform. The control logic is as follows: ; in, It is Action instructions for the region, It is the displacement increment set according to the action type; Flexible materials ensure smooth transition of deformation to avoid abrupt changes; It should be noted that by adopting a partitioned independent drive unit composed of a flexible pneumatic actuator and a shape memory alloy, and introducing new materials such as electroactive polymer (EAP) as an alternative solution, millimeter-level deformation control and smooth transition adjustment are achieved. Combined with the limit mechanism and the force control / position control dual-mode switching function, the accuracy and safety of the adjustment are further improved. The design takes into account the adjustment accuracy, response speed and patient comfort, and solves the problem of rough adjustment and easy secondary injury of traditional mechanical structures, achieving the multiple effects of realizing refined support adjustment, improving patient experience and extending the service life of the equipment.

[0027] The state feedback module is used to resample the contact state between the bed and the human body after the adjustment is completed by the execution drive module by calling the perception acquisition module again, and compare and analyze the data before and after the adjustment to output the adjustment effect evaluation result; Furthermore, the state feedback module re-samples the contact state between the bed and the human body after the adjustment by the execution drive module by calling the perception acquisition module again, and compares and analyzes the data before and after the adjustment, and outputs the adjustment effect evaluation results, including: After the adjustment is completed, the sensing acquisition module is started again to obtain new original sensor signals; Compare point by point with the data before adjustment and calculate the pressure change rate: ; Compare the temperature changes at each point , humidity changes ; The comprehensive evaluation formula is as follows: ; in, , , is the weighting coefficient, It is a comprehensive score, the smaller it is, the better the adjustment effect is; Output the adjustment effect evaluation results for the remote interaction module to upload to the medical staff terminal; It should be noted that by comparing multi-dimensional data such as pressure, temperature, and humidity before and after adjustment, and using a weighted scoring formula for comprehensive evaluation, a quantitative evaluation of the support adjustment effect is achieved. After the introduction of the image recognition auxiliary judgment mechanism, the evaluation dimensions are richer and the judgment basis is more comprehensive. The closed-loop feedback mechanism can not only be used to instantly verify the effectiveness of the adjustment strategy, but also to reversely optimize the learning process of individual modeling and strategy generation modules, forming a continuously improving intelligent system, thereby achieving the effect of improving the system's self-optimization ability and enhancing the accuracy and safety of adjustment.

[0028] The remote interaction module is used to upload the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to the remote terminal device for medical staff to view and set personalized intervention parameters; Furthermore, the remote interaction module uploads the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to the remote terminal device for medical staff to view and set personalized intervention parameters, including: local communication unit and cloud service platform; The local communication unit uses either Wi-Fi or Bluetooth protocols to connect to the bed control system and receive the following three types of data: Physiological assessment data output by the data fusion module; The supporting action instruction sequence output by the strategy generation module; The regulation effect evaluation result output by the state feedback module; The local communication unit uploads the above data to the cloud service platform via one of the MQTT or HTTP protocols; The cloud service platform categorizes and stores data and provides a visual interface to display the patient's real-time pressure distribution thermodynamic map, temperature distribution map, humidity change curve, and ambient temperature and humidity trends; Medical staff can access the cloud platform through remote terminal devices to view historical data and set personalized intervention parameters, namely the target pressure value for each area; The personalized intervention parameters are sent back to the strategy generation module as the target basis for dynamic support adjustment in the next cycle; The remote interaction module also supports abnormal warning push function. When the pressure value of a certain area exceeds the safety threshold or the regulation effect score continues to deteriorate, an alarm notification will be automatically sent to the designated terminal device. All remote operations are authenticated and data encrypted to protect patient privacy and system security; It should be noted that by using Wi-Fi or Bluetooth protocols to connect to the bed control system and uploading data to the cloud service platform through MQTT or HTTP protocols, remote data visualization and personalized intervention parameter setting are achieved. The platform supports identity authentication, data encryption and abnormal warning push functions, ensuring the security and privacy protection of the system. The remote interaction mechanism not only makes it convenient for medical staff to grasp the patient's status at any time and make intervention adjustments, but also enhances the compatibility and scalability of the system with the hospital information system, achieving the purpose of improving nursing efficiency, realizing remote management and intelligent decision-making.

[0029] This embodiment also provides a computer device, which is suitable for the case of an intelligent rehabilitation bed dynamic support system based on pressure feedback, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent rehabilitation bed dynamic support system based on pressure feedback proposed in the above embodiment.

[0030] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0031] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by the processor, it implements the intelligent rehabilitation bed dynamic support system based on pressure feedback as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0032] In summary, the present invention processes the original sensor signal by utilizing the Kalman filter algorithm, constructs the state equation and observation equation, and calculates the Kalman gain to update the state estimate, thereby obtaining a comprehensive health indicator reflecting the microcirculation status of the local tissue. The method effectively filters the noise, improves the accuracy and reliability of the data, and enables the system to more accurately identify the health status of the local tissue. A machine learning model is used to combine the patient's body shape, weight, medical history information and historical pressure distribution data to establish a personalized pressure-sensitive area model. The method can generate support feature parameters customized according to the specific situation of each patient, ensuring a high degree of personalization of the support adjustment strategy. The reinforcement learning algorithm is used to predict future high-risk areas for pressure ulcers and formulate dynamic support adjustment strategies based on the results of the data fusion module and the individual modeling module. The method allows the system to autonomously learn and optimize its adjustment strategy to adapt to the changing needs of different patients.

[0033] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Intelligent rehabilitation bed dynamic support system based on pressure feedback, characterized by: include: Perception and acquisition module, data fusion module, individual modeling module, strategy generation module, execution drive module, state feedback module, and remote interaction module; The sensing and acquisition module is used to collect the pressure, temperature, humidity and heat distribution information of the contact area between the bedridden patient and the bed in real time using a distributed multimodal sensing method, and output the original sensing signal; The data fusion module is used to perform multi-source data fusion processing on the original sensor signals from the sensing acquisition module using a Kalman filter algorithm, extract comprehensive health indicators reflecting the microcirculation status of local tissues, and output fused physiological assessment data; The individual modeling module is used to use a machine learning model to train the patient's body shape, weight, medical history information and historical pressure distribution data to establish a personalized pressure-sensitive area model and output personalized support feature parameters; The strategy generation module is used to use a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module and the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure sores, generate a dynamic support adjustment strategy, and output a support action instruction sequence; The execution drive module is used to receive the support action instruction sequence output by the strategy generation module using a partitioned independent drive unit composed of a flexible pneumatic actuator and a shape memory alloy, and implement millimeter-level deformation control on the corresponding area of ​​the bed to achieve dynamic support adjustment on demand; The state feedback module is used to resample the contact state between the bed and the human body after the adjustment is completed by the execution drive module by calling the perception acquisition module again, and compare and analyze the data before and after the adjustment to output the adjustment effect evaluation result; The remote interaction module is used to upload the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to a remote terminal device for medical staff to view and set personalized intervention parameters.

2. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 1, characterized in that: The sensing acquisition module uses a distributed multimodal sensing method to collect pressure, temperature, humidity, and heat distribution information of the contact area between the bedridden patient and the bed in real time, and outputs the original sensing signal, including: Multiple sensor nodes are arranged on the surface of the bed, each node integrates a pressure sensor, an infrared thermopile sensor, a humidity sensor and a temperature sensor; Each sensor node converts the collected physical quantity into an electrical signal and converts it into a digital signal through an A / D converter; The pressure sensor is used to measure the pressure value at each point of the contact surface between the human body and the bed; Infrared thermopile sensor is used to measure the thermal radiation intensity of the local skin surface , and converted to equivalent skin temperature; Humidity sensor is used to measure the relative humidity of the bed surface microenvironment ; Temperature sensor used to measure ambient temperature ; All collected data are packaged as raw sensor signals , and transmitted to the data fusion module for subsequent processing.

3. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 2, characterized in that: The data fusion module uses the Kalman filter algorithm to perform multi-source data fusion processing on the original sensor signals from the sensing acquisition module, extracts comprehensive health indicators reflecting the microcirculatory status of local tissues, and outputs fused physiological assessment data, including: Construct the state equation, the expression is: ； in, is the state vector, which contains the pressure, temperature, humidity and ambient temperature estimates at the current moment, is the state transition matrix, is the control input matrix, is the external disturbance input, is the process noise, which follows a normal distribution ; The observation equation is: ； in, is the observation vector, i.e., the actual sensor reading, is the observation matrix, is the observation noise, which follows a normal distribution ; The Kalman gain KkKk is calculated as follows: ； The updated state estimate is: ； Output fused physiological assessment data , which represents the pressure, temperature, humidity and ambient temperature values ​​at each point after filtering optimization.

4. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 3, characterized in that: The individual modeling module uses a machine learning model to train the patient's body shape, weight, medical history information, and historical pressure distribution data to establish a personalized pressure-sensitive area model and output personalized support feature parameters, including: Input includes basic patient information, including weight, height, age, and medical history; Simultaneously input historical pressure distribution data , indicating the past The pressure value of each sensor at a time point; Use convolutional neural network (CNN) to extract features from the spatial pressure distribution map and generate spatial feature vectors; Use the long short-term memory network LSTM to model time series data and generate time feature vectors; The spatial feature vector and the temporal feature vector are concatenated and input into the fully connected layer to output the personalized support feature parameters. ,in Indicates the The pressure sensitivity weight of each area; Output It is sent to the strategy generation module to guide subsequent dynamic support adjustments.

5. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 4, characterized in that: The strategy generation module uses a reinforcement learning algorithm to combine the physiological assessment data output by the data fusion module and the support characteristic parameters output by the individual modeling module to predict future high-risk areas for pressure ulcers, generate a dynamic support adjustment strategy, and output a support action instruction sequence, including: The reinforcement learning model uses the deep Q network DQN to define the state space Pressure distribution diagram after fusion ; The action space is defined as the direction and amplitude of deformation adjustment of each partition of the bed; The reward function is designed as follows: ； in, is the individualized support feature parameter, is the current pressure value, is the target pressure range; In each decision cycle, the current state is input and support feature parameters , output the optimal action sequence ; The supporting action instruction sequence is encoded into an instruction format that can be recognized by the execution driver module.

6. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 5, characterized in that: The execution drive module uses a partitioned independent drive unit composed of a flexible pneumatic actuator and a shape memory alloy to receive the support action instruction sequence output by the strategy generation module, and implements millimeter-level deformation control on the corresponding area of ​​the bed to achieve on-demand dynamic support adjustment, including: The bed is divided into Independent control areas, each area is equipped with a flexible pneumatic actuator and a shape memory alloy wire; The pneumatic actuator is supplied with air by a micro air pump, and the air intake / exhaust volume is controlled by a solenoid valve, thereby changing the height of the area; The shape memory alloy wire is controlled by a constant current power supply, and after being energized, a phase change occurs to generate a contraction force, thereby driving the local bed plate to deform; The control logic is as follows: ； in, It is Action instructions for the region, It is the displacement increment set according to the action type; The flexible material ensures smooth transition of deformation to avoid abrupt changes.

7. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 6, characterized in that: The state feedback module re-samples the contact state between the bed and the human body after the adjustment by the execution drive module by calling the perception acquisition module again, and compares and analyzes the data before and after the adjustment, and outputs the adjustment effect evaluation results, including: After the adjustment is completed, the sensing acquisition module is started again to obtain new original sensor signals; Compare point by point with the data before adjustment and calculate the pressure change rate: ； Compare the temperature changes at each point , humidity changes ; The comprehensive evaluation formula is as follows: ； in, , , is the weighting coefficient, It is a comprehensive score, the smaller it is, the better the adjustment effect is; Output the adjustment effect evaluation results for the remote interaction module to upload to the medical staff terminal.

8. The intelligent rehabilitation bed dynamic support system based on pressure feedback according to claim 7, characterized in that: The remote interaction module uploads the physiological assessment data output by the data fusion module, the adjustment strategy output by the strategy generation module, and the effect evaluation results output by the state feedback module to the remote terminal device for medical staff to view and set personalized intervention parameters, including: local communication unit and cloud service platform; The local communication unit is connected to the bed control system using either Wi-Fi or Bluetooth protocols to receive the following three types of data: Physiological assessment data output by the data fusion module; The supporting action instruction sequence output by the strategy generation module; The regulation effect evaluation result output by the state feedback module; The local communication unit uploads the above data to the cloud service platform via one of the MQTT or HTTP protocols; The cloud service platform categorizes and stores data and provides a visual interface to display the patient's real-time pressure distribution thermodynamic map, temperature distribution map, humidity change curve, and ambient temperature and humidity trends; Medical staff can access the cloud platform through remote terminal devices to view historical data and set personalized intervention parameters, namely the target pressure value for each area; The personalized intervention parameters are sent back to the strategy generation module as the target basis for dynamic support adjustment in the next cycle; The remote interaction module also supports abnormal warning push function. When the pressure value of a certain area exceeds the safety threshold or the regulation effect score continues to deteriorate, an alarm notification will be automatically sent to the designated terminal device. All remote operations are subject to identity authentication and data encryption to ensure patient privacy and system security.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the intelligent rehabilitation bed dynamic support system based on pressure feedback according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent rehabilitation bed dynamic support system based on pressure feedback according to any one of claims 1 to 7 are implemented.

Images