A pressure injury monitoring and early warning method and system based on a wearable device

Abstract

The application provides a pressure injury monitoring and early warning method and system based on a wearable device. The method collects pressure state parameters; performs pressure communication condition judgment according to the pressure state parameters, extracts multiple pressure communication regions, extracts pressure communication characteristics and local electrical impedance parameters corresponding to any one pressure communication region; constructs a pressure-impedance coupling feature vector according to the pressure communication characteristics and local electrical impedance information corresponding to the pressure communication region; performs feature learning based on the pressure-impedance coupling feature vector, determines a pressure response index of the pressure communication region; determines an injury early warning risk level of the corresponding pressure communication region according to the pressure response index, and sends the injury early warning risk level corresponding to each pressure communication region to a cloud data center of the wearable device. The pressure response of the pressure position can be analyzed according to the pressure-impedance coupling characteristics, and the pressure injury risk can be early warned.

Description

Technical Field

[0001] This application relates to the field of damage monitoring technology, and more specifically, to a method and system for monitoring and early warning of pressure injuries based on wearable devices. Background Technology

[0002] Pressure ulcers are a group of diseases caused by prolonged pressure and friction on the local skin, leading to impaired blood circulation and tissue ischemia and hypoxia, resulting in pressure damage to the skin and subcutaneous tissues. Clinical manifestations typically include skin redness, blisters, and ulcers. In severe cases, they can progress to deep tissue necrosis and may involve muscle and bone tissue, further leading to serious complications such as infection and even sepsis. Stage III and above severe pressure ulcers account for approximately 30% of all cases, with a mortality rate of 10%–20%.

[0003] In existing technologies, the monitoring and early warning of pressure injuries mainly rely on manual clinical observation to assess the condition. For patients at risk of pressure injuries who have not yet shown symptoms, traditional pressure ulcer risk assessment tools mainly rely on basic indicators such as the patient's mobility, nutritional status, and skin moisture for assessment. These tools are difficult to use for risk screening and identification, and their sensitivity and specificity are relatively low. They often overlook the dynamic changes in the patient's condition, resulting in insufficient risk identification and delayed prevention and control measures. Therefore, how to effectively monitor the risk of pressure injuries and improve the accuracy of early warning of pressure injury risks has become an urgent problem to be solved. Summary of the Invention

[0004] This application provides a method and system for monitoring and warning of pressure injuries based on wearable devices, which can perform pressure response analysis on the pressure-impedance coupling characteristics to achieve early warning of pressure injury risks.

[0005] In a first aspect, this application provides a method for monitoring and early warning of pressure injuries based on wearable devices. This method can be executed by a network device, or it can be executed by a chip configured in the network device. This application does not limit the method in this regard.

[0006] Specifically, the method includes: Pressure status parameters are acquired through a pressure sensor array in a wearable device. Based on the pressure state parameters, the pressure connectivity conditions are determined, and multiple pressure connectivity regions are extracted. For any pressure connectivity region, the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region are extracted. A pressure-impedance coupling feature vector is constructed based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region. Based on the pressure-impedance coupling feature vector, feature learning is performed to determine the pressure response index of the pressure-connected region. Based on the pressure response index, the damage warning risk level of the corresponding pressure connection area is determined, and the damage warning risk level of each pressure connection area is sent to the cloud data center of the wearable device.

[0007] In conjunction with the first aspect, in certain implementations of the first aspect, determining the pressure connectivity condition based on the pressure state parameters and extracting multiple pressure connectivity regions specifically includes: Based on the pressure value in the pressure state parameters and the preset pressure threshold, multiple effective pressure-bearing units are determined, and spatial adjacency relationships are established according to the position coordinates of each effective pressure-bearing unit. Effective pressure-bearing units that meet the spatial adjacency condition are divided into the same pressure region. Based on the pressure time corresponding to each pressure region, the region is filtered to obtain multiple pressure-connected regions.

[0008] In conjunction with the first aspect, in certain implementations of the first aspect, extracting the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region specifically includes: The pressure value and spatial location information of each pressure sensing unit in the pressure connectivity area are obtained, and the pressure connectivity features are extracted based on the pressure value and spatial location information. The corresponding impedance detection area is determined based on the spatial location of the pressure connection area. The impedance parameters corresponding to the impedance detection area are collected by the impedance detection electrode array to obtain the local impedance parameters.

[0009] In conjunction with the first aspect, in certain implementations of the first aspect, constructing a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region specifically includes: The local impedance parameters corresponding to the pressure connection region are dynamically monitored, the starting time of the change in local impedance value is extracted, and the starting time of the change is determined as the response starting time. Using the response start time as a reference, the pressure connectivity features and local impedance parameters are time-aligned to obtain the aligned pressure feature sequence and impedance feature sequence. A pressure-impedance coupling feature vector is constructed based on the response amplitude characteristics, response delay characteristics, response correlation characteristics, and attenuation trend characteristics between the pressure feature sequence and the impedance feature sequence.

[0010] In conjunction with the first aspect, in certain implementations of the first aspect, determining the damage warning risk level of the corresponding pressure-connected region based on the pressure response index specifically includes: Obtain a baseline response index, and determine the response decay characteristics based on the pressure response index and the baseline response index; Based on the aforementioned response attenuation characteristics, a risk level mapping is performed to determine the corresponding damage warning risk level.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, during the process of collecting pressure state parameters through a pressure sensing array in a wearable device, the pressure state parameters include the pressure value and its corresponding location coordinate information and sampling time.

[0012] In conjunction with the first aspect, in some implementations of the first aspect, the wearable device includes a flexible pressure sensing array.

[0013] Secondly, this application provides a pressure injury monitoring and early warning system based on a wearable device, which includes a risk identification unit, the risk identification unit comprising: Pressure acquisition module, used to acquire pressure status parameters; The feature recognition module is used to determine the pressure connectivity condition based on the pressure state parameters, extract multiple pressure connectivity regions, and extract the pressure connectivity features and local impedance parameters corresponding to any pressure connectivity region. The data processing module is used to construct a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region. The data processing module also performs feature learning based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure-connected region. The risk classification module determines the damage warning risk level of the corresponding pressure connection area based on the pressure response index, and sends the damage warning risk level of each pressure connection area to the cloud data center of the wearable device.

[0014] Thirdly, this application provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the aforementioned method for monitoring and warning of pressure injuries based on a wearable device.

[0015] Fourthly, this application provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to perform the operations described above in the pressure injury monitoring and early warning method based on a wearable device.

[0016] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: This application provides a pressure injury monitoring and early warning method and system based on wearable devices. First, pressure state parameters are collected using a pressure sensor array in the wearable device. Based on the pressure state parameters, pressure connectivity conditions are determined, and multiple pressure connectivity regions are extracted. For any given pressure connectivity region, pressure connectivity features and local impedance parameters are extracted. A pressure-impedance coupling feature vector is constructed based on the pressure connectivity features and local impedance information. Feature learning is performed based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure connectivity region. The damage early warning risk level of the corresponding pressure connectivity region is determined based on the pressure response index, and the damage early warning risk levels corresponding to each pressure connectivity region are sent to the cloud data center of the wearable device.

[0017] Therefore, this application acquires pressure sensor array data from wearable devices, identifies pressure-connected regions, and constructs pressure-impedance coupling features by combining local electrical impedance parameters. This enables dynamic characterization of tissue microcirculation changes in the pressure-affected area under continuous pressure. Specifically, through the synergistic analysis of pressure and impedance information, it captures the trend of perfusion changes within the tissue in the early stages of pressure application. By constructing pressure response indicators, it quantifies the sensitivity of blood perfusion changes under unit pressure, thereby distinguishing the response differences of different individuals and different tissue regions under the same pressure conditions. This improves the individual adaptability and precision of pressure injury risk assessment. Based on pressure-impedance coupling features, feature learning is performed, and combined with response attenuation characteristics, damage warning risk level classification is achieved. This transforms pressure injury risk assessment from the traditional static pressure judgment mode to a prediction mode based on dynamic physiological response, realizing early identification and graded warning of pressure injury risk.

[0018] In summary, this application can perform pressure response analysis on the pressure-impedance coupling characteristics, thereby enabling early warning of pressure damage risks. Attached Figure Description

[0019] Figure 1 This is an exemplary flowchart illustrating a pressure injury monitoring and early warning method based on a wearable device, according to some embodiments of this application. Figure 2 This is a schematic diagram of the risk identification unit according to some embodiments of this application; Figure 3 This is a schematic diagram of the structure of a computer terminal device that implements a pressure injury monitoring and early warning method based on a wearable device, according to some embodiments of this application. Detailed Implementation

[0020] This application collects pressure state parameters using a pressure sensor array in a wearable device; it determines pressure connectivity conditions based on these parameters, extracts multiple pressure connectivity regions, and for any given region, extracts its corresponding pressure connectivity features and local impedance parameters; it constructs a pressure-impedance coupling feature vector based on these features and impedance information; it performs feature learning based on this vector to determine the pressure response index of the pressure connectivity region; it determines the damage warning risk level of the corresponding pressure connectivity region based on the pressure response index, and sends the damage warning risk levels for each pressure connectivity region to the cloud data center of the wearable device. This enables pressure response analysis of the pressure location based on pressure-impedance coupling characteristics, achieving early warning of pressure-related damage risks.

[0021] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart of a pressure injury monitoring and early warning method based on a wearable device, according to some embodiments of this application. The pressure injury monitoring and early warning method 100 based on a wearable device mainly includes the following steps: In step S101, pressure state parameters are collected through the pressure sensor array in the wearable device.

[0022] It should be noted that the wearable device described in this application includes a flexible pressure sensing array. The flexible pressure sensing array is formed by multiple pressure sensing units distributed according to a preset array pattern and is attached to the skin surface of the target object. The pressure sensing units sense the pressure state of each contact area of ​​the target object's body in real time. Specifically, each pressure sensing unit collects the pressure signal at the corresponding position according to a preset sampling period and converts the collected pressure signal into digital pressure data to form a pressure distribution matrix corresponding to the current moment. The pressure state parameters include the pressure value and the position coordinate information corresponding to the pressure value. The position coordinate information is used to characterize the spatial position of the pressure sensing unit in the pressure sensing array.

[0023] Optionally, in some embodiments, during the process of acquiring pressure state parameters through a pressure sensing array in a wearable device, the pressure state parameters include the pressure value and its corresponding location coordinate information and sampling time.

[0024] In step S102, pressure connectivity conditions are determined based on the pressure state parameters, and multiple pressure connectivity regions are extracted. For any pressure connectivity region, the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region are extracted. Optionally, in some embodiments, determining pressure connectivity conditions based on the pressure state parameters and extracting multiple pressure connectivity regions specifically includes: Based on the pressure value in the pressure state parameters and the preset pressure threshold, multiple effective pressure-bearing units are determined, and spatial adjacency relationships are established according to the position coordinates of each effective pressure-bearing unit. Effective pressure-bearing units that meet the spatial adjacency condition are divided into the same pressure region. Based on the pressure time corresponding to each pressure region, the region is filtered to obtain multiple pressure-connected regions.

[0025] In specific implementation, the pressure sensor array has multiple pressure sensor units. Based on the pressure value in the pressure state parameters and a preset pressure threshold, it is determined whether each pressure sensor unit belongs to an effective pressure-bearing unit. Among them, the unit with a pressure value greater than or equal to the preset pressure threshold is identified as an effective pressure-bearing unit. For all effective pressure-bearing units, a spatial adjacency relationship is established based on their position coordinates in the pressure sensor array. In this application, the spatial adjacency relationship can adopt a common proximity judgment method, such as connecting each pressure unit with its upper, lower, left, right, and diagonally adjacent units to determine the continuous regions that can be formed between the units. Multiple effective pressure-bearing units that meet the spatial adjacency conditions are grouped into the same pressure region. Each pressure region includes several adjacent and effective pressure units forming a connected pressure-bearing region. For each initial pressure region, its corresponding continuous pressure time is calculated, that is, the length of time that the pressure value continues to exceed the threshold. In some embodiments, the pressure uniformity parameter can also be calculated simultaneously to determine the average value and standard deviation of the pressure value in the region to measure the pressure uniformity of the region. Regions where the continuous pressure time reaches the preset duration threshold and the pressure uniformity parameter meets the preset conditions are retained to obtain multiple pressure connected regions.

[0026] In some embodiments, extracting the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region specifically includes: The pressure value and spatial location information of each pressure sensing unit in the pressure connectivity area are obtained, and the pressure connectivity features are extracted based on the pressure value and spatial location information. The corresponding impedance detection area is determined based on the spatial location of the pressure connection area. The impedance parameters corresponding to the impedance detection area are collected by the impedance detection electrode array to obtain the local impedance parameters.

[0027] In specific implementation, for any pressure connectivity region, at each sampling time, a set of pressure values ​​corresponding to all pressure detection units constituting the pressure connectivity region is acquired, and the average pressure, peak pressure, and pressure gradient are calculated based on the pressure value set. The average pressure is obtained by averaging the pressure values ​​corresponding to all pressure detection units within the pressure connectivity region; the peak pressure is obtained by extracting the maximum pressure value from the pressure value set; the pressure gradient is obtained by calculating and averaging the pressure differences between adjacent pressure detection units, used to identify the uniformity of pressure distribution within the pressure connectivity region; the connectivity area is calculated based on the number of pressure detection units contained within the pressure connectivity region and the actual coverage area corresponding to a single pressure detection unit; the continuous pressure duration is obtained by accumulating the duration for which the pressure connectivity region continuously meets the pressure threshold condition; the pressure fluctuation characteristic is obtained by calculating the pressure standard deviation based on the average pressure changes corresponding to multiple consecutive sampling times within a preset time window, used to characterize the stability of the pressure state; after obtaining the pressure connectivity feature vectors corresponding to multiple sampling times, they are arranged according to the sampling time order to form a pressure connectivity feature matrix, which serves as the pressure connectivity feature. Each row of the pressure connectivity feature matrix corresponds to a pressure connectivity feature vector at a sampling time, and each column corresponds to a pressure connectivity feature parameter.

[0028] In some embodiments, during the process of acquiring the impedance parameters corresponding to the impedance detection area through an impedance detection electrode array to obtain local impedance parameters, at each sampling time, the impedance value set of each impedance detection electrode in the corresponding impedance detection area is obtained, and a local impedance feature vector is constructed based on the impedance value set. The local impedance feature vector includes the impedance value, impedance change, impedance change rate, and impedance uniformity index corresponding to the current time. The current impedance value is obtained by averaging the impedance values ​​corresponding to all impedance detection electrodes in the impedance detection area; the impedance change is determined by the difference between the current impedance value and a preset reference impedance; the impedance change rate is calculated by the ratio of the impedance change between adjacent sampling times to the corresponding time interval; and the impedance uniformity index is obtained by calculating the standard deviation of the impedance values ​​corresponding to each impedance detection electrode in the impedance detection area. After obtaining the local impedance feature vectors corresponding to multiple sampling times, they are arranged in the order of sampling time to form a local impedance parameter in the form of a feature matrix. Each row of this feature matrix corresponds to a local impedance feature vector at a sampling time, and each column corresponds to a local impedance feature parameter.

[0029] It should be noted that when local pressure exceeds the capillary closure pressure (usually about 32 mmHg) and persists, it will compress the capillaries, leading to a reduction in local blood flow and changes in the conductivity of the tissue. The pressure connectivity feature matrix described in this application can effectively characterize the pressure condition of the pressure connectivity area under external mechanical load, and the local electrical impedance parameter matrix can reflect the dynamic changes in the internal tissue state of the area. Therefore, through coupled feature analysis and feature learning of the pressure connectivity feature matrix and the local electrical impedance parameter matrix, personalized assessment of microcirculation changes in the pressure area can be achieved, taking into account individual physiological differences. This can identify the changing trend of the blood perfusion index, thereby predicting the latent risk of pressure injury in advance, which helps to intervene in a timely manner and reduce the possibility of pressure injury for wearable device users.

[0030] In some embodiments, the wearable device includes a flexible pressure-resistance shared electrode array, formed of a flexible conductive material and disposed on the side of the wearable device that contacts the user's skin, for simultaneously acquiring pressure state parameters and local tissue impedance parameters. In pressure detection mode, each electrode unit of the flexible electrode array undergoes a slight deformation under external force, causing a change in the piezoresistive value of the electrode unit. The pressure value at the corresponding location is acquired by reading the electrical signal of the electrode unit. In impedance detection mode, the flexible electrode array applies a weak excitation between the electrodes and measures the voltage response of the tested area, thereby obtaining the local tissue impedance parameters of the corresponding pressure-connected region.

[0031] In specific implementation, for any pressure-connected region, after identifying its location and range using pressure detection mode, the system switches to impedance detection mode. Only the common electrode corresponding to the pressure-connected region is measured for impedance, thereby reducing power consumption and data volume. A pressure connectivity feature vector and a local impedance feature vector are constructed for each sampling moment. The pressure connectivity feature vector includes average pressure, peak pressure, pressure gradient, connectivity area, duration of pressure application, and pressure fluctuation characteristics. The local impedance feature vector includes the current impedance value, impedance change, impedance change rate, and impedance uniformity index. By collecting these feature vectors at continuous sampling moments, a pressure connectivity feature matrix and a local impedance parameter matrix are formed.

[0032] Optionally, in some embodiments, the flexible pressure-impedance shared electrode array may employ a multilayer flexible thin film structure, including at least one layer for pressure sensing and at least one layer for impedance measurement. In some other embodiments, pressure and impedance may be measured separately in different modes using the same flexible electrode unit, which is not limited in this application.

[0033] In step S103, a pressure-impedance coupling feature vector is constructed based on the pressure connectivity features and local impedance information corresponding to the pressure connectivity region.

[0034] Optionally, in some embodiments, constructing a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region specifically includes: The local impedance parameters corresponding to the pressure connection region are dynamically monitored, the starting time of the change in local impedance value is extracted, and the starting time of the change is determined as the response starting time. Using the response start time as a reference, the pressure connectivity features and local impedance parameters are time-aligned to obtain the aligned pressure feature sequence and impedance feature sequence. A pressure-impedance coupling feature vector is constructed based on the response amplitude characteristics, response delay characteristics, response correlation characteristics, and attenuation trend characteristics between the pressure feature sequence and the impedance feature sequence.

[0035] In practice, local impedance parameters are dynamically monitored, and impedance offset is calculated based on the impedance values ​​corresponding to continuous sampling times. The impedance offset characterizes the degree of change of the current impedance value relative to a reference impedance value. The reference impedance value can be represented by the average impedance value within a preset time window before entering a state of continuous pressure. For any sampling time, the corresponding impedance offset is obtained by calculating the difference between the current impedance value and the reference impedance value, and the impedance offset is compared with a preset offset threshold. When the impedance offset is higher than the preset offset threshold for multiple consecutive sampling periods, it is determined that the local tissue has responded to continuous pressure, and the sampling time that first meets the condition is determined as the impedance change initiation time, which is also determined as the response initiation time. It should be noted that, in order to eliminate time-series biases caused by individual differences and sampling delays, and to reduce the impact of individual differences on subsequent analysis results, this application uses the moment of obvious physiological response as the starting point for analysis of data corresponding to different individuals, different monitoring areas, and different monitoring cycles. Specifically, the data point corresponding to the response start moment is defined as a unified time zero point, and the time coordinate system is reconstructed according to the time order before and after the response start moment, thereby obtaining the aligned pressure feature sequence and impedance feature sequence. Then, coupling relationship features are extracted from the aligned pressure feature sequence and impedance feature sequence to construct a pressure-impedance coupling feature vector. Optionally, in some embodiments, the coupling relationship features in the pressure-impedance coupling feature vector include: response amplitude features, response delay features, response correlation features, and attenuation trend features. Specifically, the response amplitude feature is used to characterize the intensity of the influence of pressure change on impedance change. Multiple consecutive sampling times after the response start time are selected, and the changes in pressure connectivity features and local impedance parameters relative to the response start time are calculated respectively. The impedance change is then compared with the pressure change to obtain the impedance response intensity corresponding to a unit pressure change, thereby determining the response amplitude feature. The response delay feature is used to characterize the time lag between pressure action and impedance response, obtained based on the time interval between the start of continuous pressure and the start of impedance change. The response correlation feature is used to characterize the correlation between local impedance parameters and the cumulative pressure after the response start time. Specifically, the cumulative pressure sequence and local impedance parameter sequence after the impedance change start time are obtained, and normalized based on the maximum value of the sequence, obtaining the absolute value of the correlation coefficient between the sequences as the response correlation feature. The attenuation trend relationship is used to characterize the rate of deterioration of tissue state under continuous pressure conditions, determined by extracting the average rate of change of impedance parameters after the response start time.

[0036] In some embodiments, the local impedance parameter can be determined based on the regional average impedance value at different sampling times in the pressure communication region. Specifically, by arithmetically averaging the real-time impedance values ​​of all impedance detection units within the pressure communication region, a sequence of regional average impedance parameters corresponding to each sampling time is obtained, reflecting the overall impedance level change of the region under continuous pressure. In other embodiments, the local impedance parameter can be further combined with the impedance change, impedance change rate, and impedance uniformity index in the local impedance feature vector as correction weighting coefficients to weight and correct the regional average impedance value, thereby obtaining a more reliable and stable local impedance parameter. Specifically, for each sampling time, the regional average impedance value at that time can be calculated first. Based on the local dynamic amplitude reflected by the impedance change, the response speed reflected by the impedance change rate, and the spatial distribution consistency reflected by the impedance uniformity index, correction weighting coefficients are generated. By multiplying the regional average impedance value by the correction weighting coefficients, the corrected local impedance parameter is obtained.

[0037] In step S104, feature learning is performed based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure-connected region.

[0038] It should be noted that the pressure response index in this application is used to quantify the ability of the blood perfusion index to respond to changes under continuous pressure, i.e., the sensitivity of tissue to changes in the blood perfusion index under pressure. Its unit is ml / min / kPa. The pressure response index is obtained through feature learning of the pressure-impedance coupling characteristics under pressure conditions. Specifically, when the skin and soft tissue of a specific part of the human body are subjected to continuous external pressure, the local tissue in that area is subjected to mechanical compression, leading to deformation of the microvascular bed and a decrease in local microcirculation perfusion level. As the blood perfusion level decreases, the balance between oxygen metabolism and fluid exchange in the local tissue is disrupted, further leading to remodeling of interstitial fluid distribution, changes in extracellular fluid content, and measurable dynamic changes in tissue electrical impedance characteristics.

[0039] Since tissue electrical impedance is correlated with its water content, extracellular fluid ratio, and blood perfusion status, dynamic monitoring of local electrical impedance parameters can indirectly reflect changes in local blood perfusion. Based on this, this application constructs pressure-impedance coupling features to identify the dynamic relationship between pressure intensity and impedance response changes, and models this relationship using feature learning to obtain a pressure response index characterizing the ability to respond to changes in blood perfusion under unit pressure. This index reflects the microcirculation regulation capacity and perfusion reduction sensitivity of local tissues under continuous pressure, providing a quantitative basis for early identification and graded warning of pressure injury risk.

[0040] Optionally, in some embodiments, a multi-hidden-layer feedforward neural network model is used to learn features from the pressure-impedance coupling feature vector to determine the pressure response index of the pressure-connected region. A specific embodiment of this application for determining the pressure response index of the pressure-connected region is given below: The multi-hidden-layer feedforward neural network model includes an input layer, at least two hidden layers, and an output layer. The input layer is used to receive the pressure-impedance coupling feature vector, the hidden layer is used to perform nonlinear mapping and feature abstraction on the input features, and the output layer is used to output the pressure response index. The pressure-impedance coupling feature vector includes response amplitude features, response delay features, response correlation features, and attenuation trend features, and serves as the feature input of the neural network model.

[0041] During the model training phase, a training sample set is constructed, which is derived from the pressure response experimental data of actual healthy experimental populations. During the experiment, a controllable continuous pressure is applied to the target pressure area of ​​the experimental subjects, and the cumulative pressure per unit area is recorded. Pressure state parameters and local impedance parameters are collected synchronously through wearable devices, and a pressure-impedance coupling feature sequence is generated according to a preset sampling interval. At the same time, a blood perfusion monitoring device (such as a laser Doppler blood flow detection device) is used to obtain the blood perfusion index of the corresponding pressure area before and after pressure, and the ratio between the blood perfusion change ratio and the cumulative pressure is calculated as a supervised learning label. The blood perfusion change ratio is used to characterize the degree of decrease in blood perfusion relative to the baseline state after pressure.

[0042] The training samples are input into a multi-hidden-layer feedforward neural network model for training. The network parameters are iteratively optimized using a backpropagation algorithm. The optimization objective is to minimize the error function between the model's output pressure response index and the actual blood perfusion change ratio. In some embodiments, the error function type is the mean squared error function. During training, when the error between the model output and the actual blood perfusion change ratio exceeds a preset threshold, the model is updated by adjusting the weight parameters, bias parameters, and activation function parameters of the hidden layer neurons to improve the model's ability to fit the mapping relationship between pressure-impedance coupling features and blood perfusion changes. After the model training is completed, the pressure-impedance coupling feature vector corresponding to the pressure connectivity region acquired in real time is input into the trained multi-hidden-layer feedforward neural network model, and the corresponding pressure response index is output by the output layer.

[0043] In step S105, the damage warning risk level of the corresponding pressure connection area is determined according to the pressure response index, and the damage warning risk level of each pressure connection area is sent to the cloud data center of the wearable device.

[0044] It should be noted that the technical basis of the pressure injury monitoring and early warning method proposed in this application lies in the coupling relationship between pressure, microcirculation, and electrical characteristics of local tissues under continuous pressure. That is, when a specific part of the human body is subjected to continuous external pressure, the capillaries and microvascular beds in the skin and soft tissues of that area will undergo varying degrees of deformation, resulting in a decrease in local blood perfusion. Changes in blood perfusion will further affect the exchange process of oxygen and metabolic products in the tissue, causing changes in the distribution of tissue fluid, extracellular fluid content, and local ion environment. External identification is achieved through impedance characteristic detection. In other words, since there is a correlation between tissue impedance and tissue water content, extracellular fluid ratio, and microcirculation perfusion state, changes in local blood perfusion will cause corresponding changes in measurable impedance parameters. By dynamically monitoring local impedance parameters, the changes in the local tissue microcirculation state under pressure can be indirectly reflected, thereby achieving the characterization of changes in blood perfusion.

[0045] As can be seen, this application identifies pressure-connected regions and simultaneously acquires the pressure characteristics and local electrical impedance characteristics of these regions to construct a pressure-impedance coupling relationship, which is used to characterize the dynamic mapping relationship between external mechanical loads and the internal tissue microcirculation response. By performing feature learning on this coupling relationship, the response amplitude, response delay, and trend of blood perfusion changes under pressure can be extracted, thereby quantifying the sensitivity of different tissue regions to the decrease in perfusion under continuous pressure. Furthermore, due to differences in physiological state and microcirculation compensatory capacity, different individuals and different tissue regions exhibit different blood perfusion response characteristics under the same or similar pressure conditions. Therefore, by comparing and analyzing the perfusion pressure response characteristics of the target pressure-connected region with the response characteristics of a healthy reference region, this application can identify abnormal regions with significantly enhanced sensitivity to the decrease in blood perfusion, thereby characterizing the potential risk level of pressure injury.

[0046] Optionally, in some embodiments, determining the damage warning risk level of the corresponding pressure-connected area based on the pressure response index specifically includes: Obtain a baseline response index, and determine the response decay characteristics based on the pressure response index and the baseline response index; Based on the aforementioned response attenuation characteristics, a risk level mapping is performed to determine the corresponding damage warning risk level.

[0047] In specific implementation, the benchmark response index can be obtained through feature learning of the pressure-impedance coupling characteristics of the non-pressure-connected region under low-pressure conditions, thereby reflecting the dynamic response capability of local blood perfusion of an individual under conditions without significant pressure. In some embodiments, when conditions permit, controllable continuous pressure can be applied to the low-risk area of ​​the wearable device, and blood perfusion monitoring equipment (e.g., laser Doppler blood flow detection equipment) can be used to obtain the blood perfusion index of the corresponding pressure area before and after pressure. The benchmark response index is obtained by calculating the ratio between the proportion of blood perfusion change and the cumulative pressure. Based on the benchmark response index, the pressure response index obtained by the pressure-connected region under pressure conditions is compared with the benchmark index. By calculating the attenuation amplitude, the response attenuation characteristics of the pressure-connected region are determined. The response attenuation characteristics are used to quantify the degree of decrease in the response sensitivity of the blood perfusion index of the region under continuous pressure conditions.

[0048] It should be noted that, in this application, the response decay characteristic refers to the decrease in the sensitivity of local tissue blood perfusion indicators to the decrease in pressure over time under continuous pressure. Its magnitude reflects the tissue's tolerance to continuous pressure. Areas with weaker tissue tolerance and more easily obstructed microcirculation experience a greater decrease in blood perfusion and a faster decay rate; while areas with healthy tissue and good microcirculation function experience a smaller decrease in perfusion and a slower decay rate. Therefore, the response decay characteristic can reflect the physiological vulnerability of local tissues under pressure conditions. In some embodiments, based on extensive experimental data and clinical statistical results, different response decay characteristics can be mapped to risk levels for pressure injury. For example, areas with smaller response decay amplitudes correspond to low risk; areas with moderate decay amplitudes correspond to medium risk. By using predefined threshold intervals and establishing a mapping table, the quantified response decay characteristics can be directly converted into injury warning risk levels, thereby achieving quantitative and repeatable risk assessment.

[0049] Furthermore, the risk levels of each pressure-connected area can be comprehensively ranked, and combined with historical pressure records, area priorities, and nursing resource information, personalized nursing strategies can be generated, including adjustments to the turning frequency, local pressure intervention, and reminder notifications. This enables real-time early warning and protective measures for different risk levels, thereby improving the accuracy and timeliness of pressure injury protection for wearable device users.

[0050] Optionally, in some embodiments, after obtaining the damage warning risk level corresponding to each pressure connection area, the risk levels of each pressure connection area are uniformly summarized, and a regional risk data structure is generated. Specifically, the identification information, spatial location information, current pressure state parameters, local impedance parameters, pressure response indicators, and corresponding damage warning risk level of each pressure connection area can be associated and bound to form a multi-dimensional risk data unit, thereby constructing a risk data set containing multiple pressure connection areas. During the data upload process, the wearable device establishes a wireless communication connection with the cloud data center through its built-in communication module, which can be a cellular communication module, a Wi-Fi module, or a Bluetooth gateway forwarding module. After the connection is established, the risk data set can be encapsulated according to a preset data frame format, wherein the data frame includes at least a device identification field, a timestamp field, a pressure connection area number field, and a corresponding risk level field, and the data frame is compressed to improve transmission efficiency and data security.

[0051] In some embodiments, to reduce communication load, wearable devices trigger data uploads only when a change in risk level is detected; or they upload data in batches according to a preset time period, thereby reducing communication frequency and energy consumption. Simultaneously, different upload priorities can be set for different risk levels; for example, data corresponding to high-risk areas has higher priority. On the cloud data center side, after receiving the risk data set, the risk levels of each pressure-connected area are analyzed, and data is fused and stored based on device identification and historical monitoring data to form a continuous time-dimensional risk evolution sequence. Furthermore, trend analysis of risk level changes at multiple time points can be performed in the cloud to identify areas of rising risk and generate dynamic risk distribution maps and risk heat maps for remote monitoring and nursing decision support.

[0052] In some alternative implementations, the cloud data center can also trigger corresponding feedback mechanisms based on different risk levels. For example, when a high-risk area is detected, a reminder instruction is sent to the nursing terminal of the wearable device to prompt for turning over, pressure relief, and local nursing operations, thereby realizing cloud-based closed-loop pressure injury risk warning and control.

[0053] Therefore, this application achieves a quantitative description of the local tissue microcirculation pressure response process by coupling pressure characteristics and local electrical impedance characteristics, and realizes early identification and graded warning of pressure injury risk based on the difference in response sensitivity, thereby improving the accuracy and foresight of pressure injury monitoring.

[0054] Furthermore, in another aspect of this application, in some embodiments, this application provides a pressure injury monitoring and early warning system based on a wearable device, the system including a risk identification unit, referenced... Figure 2 The figure is a schematic diagram of the exemplary hardware and / or software structure of a risk identification unit according to some embodiments of this application. The risk identification unit 200 includes: a pressure acquisition module 201, a feature recognition module 202, a data processing module 203, and a risk classification module 204, which are described below: Pressure acquisition module 201 is used to acquire pressure status parameters; The feature recognition module 202 is used to determine the pressure connectivity condition based on the pressure state parameters, extract multiple pressure connectivity regions, and extract the pressure connectivity features and local impedance parameters corresponding to any pressure connectivity region. Data processing module 203 is used to construct a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region; The data processing module 203 also performs feature learning based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure connection region. The risk classification module 204 determines the damage warning risk level of the corresponding pressure connection area based on the pressure response index, and sends the damage warning risk level of each pressure connection area to the cloud data center of the wearable device.

[0055] The foregoing detailed an example of a pressure injury monitoring and early warning method and system based on a wearable device provided in the embodiments of this application. It is understood that, in order to achieve the above functions, the corresponding device includes hardware structures and / or software modules for performing each function.

[0056] Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in a manner that drives hardware or computer software depends on the specific application and design constraints of the technical solution. Therefore, those skilled in the art can use different methods to implement the described function for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0057] In addition, this application also provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for monitoring and warning of pressure injuries based on a wearable device.

[0058] In some embodiments, reference Figure 3The figure is a schematic diagram of the structure of a computer terminal device implementing a pressure injury monitoring and early warning method based on a wearable device, according to some embodiments of this application. The pressure injury monitoring and early warning method based on a wearable device in the above embodiments can... Figure 3 The computer terminal device 300 shown is used to implement this, and the computer terminal device 300 includes at least one communication bus 301, communication interface 302, processor 303 and memory 304.

[0059] The processor 303 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of a pressure injury monitoring and early warning method based on a wearable device in this application.

[0060] The communication bus 301 may include a path for transmitting information between the aforementioned components.

[0061] Memory 304 may be read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions, random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 304 may exist independently and be connected to processor 303 via communication bus 301. Memory 304 may also be integrated with processor 303.

[0062] The memory 304 stores program code that executes the scheme of this application, and its execution is controlled by the processor 303. The processor 303 executes the program code stored in the memory 304. The program code may include one or more software modules. In the above embodiments, the determination of the pressure response index can be achieved by the processor 303 and one or more software modules in the program code in the memory 304.

[0063] Communication interface 302 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0064] Optionally, the computer terminal device 300 may also include a power supply 305 for providing power to various devices or circuits in the real-time computer terminal device.

[0065] In a specific implementation, as one example, a computer terminal device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0066] The aforementioned computer terminal device can be a general-purpose computer terminal device or a dedicated computer terminal device. In specific implementations, the computer terminal device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer terminal device.

[0067] In addition, other aspects of this application provide a computer-readable storage medium storing at least one computer program loaded and executed by a processor to perform the operations described above in a pressure injury monitoring and early warning method based on a wearable device.

[0068] In summary, the pressure injury monitoring and early warning method and system based on wearable devices disclosed in this application collects pressure state parameters through a pressure sensor array in the wearable device; determines pressure connectivity conditions based on the pressure state parameters, extracts multiple pressure connectivity regions, and for any pressure connectivity region, extracts the corresponding pressure connectivity features and local impedance parameters; constructs a pressure-impedance coupling feature vector based on the pressure connectivity features and local impedance information of the pressure connectivity region; performs feature learning based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure connectivity region; determines the damage early warning risk level of the corresponding pressure connectivity region based on the pressure response index, and sends the damage early warning risk level corresponding to each pressure connectivity region to the cloud data center of the wearable device. This enables pressure response analysis of the pressure location based on the pressure-impedance coupling characteristics, achieving early warning of pressure injury risk.

[0069] The above descriptions are merely embodiments of this application, and common knowledge such as specific technical solutions or characteristics in the solutions are not described in detail here. It should be noted that those skilled in the art can make several modifications and improvements without departing from the technical solutions of this application, and these should also be considered within the scope of protection of this application, without affecting the effectiveness of the implementation of this application or the practicality of the patent.

[0070] The scope of protection claimed in this application shall be determined by the content of its claims. The specific embodiments described in the specification can be used to interpret the content of the claims. Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A method for monitoring and early warning of pressure injuries based on wearable devices, characterized in that, include: Pressure status parameters are acquired through a pressure sensor array in a wearable device. Based on the pressure state parameters, the pressure connectivity conditions are determined, and multiple pressure connectivity regions are extracted. For any pressure connectivity region, the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region are extracted. A pressure-impedance coupling feature vector is constructed based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region. Based on the pressure-impedance coupling feature vector, feature learning is performed to determine the pressure response index of the pressure-connected region. Based on the pressure response index, the damage warning risk level of the corresponding pressure connection area is determined, and the damage warning risk level of each pressure connection area is sent to the cloud data center of the wearable device. Specifically, determining the pressure connectivity conditions based on the pressure state parameters and extracting multiple pressure connectivity regions includes: Based on the pressure value in the pressure state parameters and the preset pressure threshold, multiple effective pressure-bearing units are determined, and spatial adjacency relationships are established according to the position coordinates of each effective pressure-bearing unit. Effective pressure-bearing units that meet the spatial adjacency condition are divided into the same pressure region. Based on the pressure time corresponding to each pressure region, the region is filtered to obtain multiple pressure-connected regions.

2. The method as described in claim 1, characterized in that, Extracting the pressure connectivity features and local impedance parameters corresponding to the pressure connectivity region specifically includes: The pressure value and spatial location information of each pressure sensing unit in the pressure connectivity area are obtained, and the pressure connectivity features are extracted based on the pressure value and spatial location information. The corresponding impedance detection area is determined based on the spatial location of the pressure connection area. The impedance parameters corresponding to the impedance detection area are collected by the impedance detection electrode array to obtain the local impedance parameters.

3. The method as described in claim 1, characterized in that, Constructing a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region specifically includes: The local impedance parameters corresponding to the pressure connection region are dynamically monitored, the starting time of the change in local impedance value is extracted, and the starting time of the change is determined as the response starting time. Using the response start time as a reference, the pressure connectivity features and local impedance parameters are time-aligned to obtain the aligned pressure feature sequence and impedance feature sequence. A pressure-impedance coupling feature vector is constructed based on the response amplitude characteristics, response delay characteristics, response correlation characteristics, and attenuation trend characteristics between the pressure feature sequence and the impedance feature sequence.

4. The method as described in claim 1, characterized in that, Determining the damage warning risk level of the corresponding pressure-connected area based on the aforementioned pressure response index specifically includes: Obtain a baseline response index, and determine the response decay characteristics based on the pressure response index and the baseline response index; Based on the aforementioned response attenuation characteristics, a risk level mapping is performed to determine the corresponding damage warning risk level.

5. The method as described in claim 1, characterized in that, During the process of collecting pressure state parameters through a pressure sensor array in a wearable device, the pressure state parameters include the pressure value, its corresponding location coordinates, and the sampling time.

6. The method as described in claim 1, characterized in that, The wearable device includes a flexible pressure sensor array.

7. A pressure injury monitoring and early warning system based on wearable devices, comprising a risk identification unit, wherein the risk identification unit is used to execute the pressure injury monitoring and early warning method based on wearable devices according to any one of claims 1 to 6, characterized in that, The risk identification unit includes: Pressure acquisition module, used to acquire pressure status parameters; The feature recognition module is used to determine the pressure connectivity condition based on the pressure state parameters, extract multiple pressure connectivity regions, and extract the pressure connectivity features and local impedance parameters corresponding to any pressure connectivity region. The data processing module is used to construct a pressure-impedance coupling feature vector based on the pressure connectivity characteristics and local impedance information corresponding to the pressure connectivity region. The data processing module also performs feature learning based on the pressure-impedance coupling feature vector to determine the pressure response index of the pressure-connected region. The risk classification module determines the damage warning risk level of the corresponding pressure connection area based on the pressure response index, and sends the damage warning risk level of each pressure connection area to the cloud data center of the wearable device.

8. A computer terminal device, characterized in that, The computer terminal device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code, such that when the code is executed in the processor, the method as described in any one of claims 1-6 is implemented.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on a computer, cause the method as described in any one of claims 1-6 to be implemented.

Images