Nursing method and system based on hierarchical monitoring mode

Abstract

The invention discloses a nursing method and system based on a hierarchical monitoring mode, and relates to the technical field of monitoring nursing, and the method comprises the following steps: firstly, according to a preset detection period, carrying out the regular detection of the skin microcirculation perfusion state of a severely burnt patient, and evaluating the tissue perfusion and wound recovery conditions; by monitoring skin microcirculation changes in real time and combining a deep learning model, the scheme accurately captures early characteristics of blood hypercoagulability trend, such as microcirculation shear stress and pulsation strength changes, generates blood flow indexes, compares the blood flow indexes with reference thresholds, early recognizes the hypercoagulability state and timely intervenes; the risks of local necrosis, infection, septicemia and the like are reduced. Based on intelligent division, patients are divided into blood rheological balance and blood hypercoagulability trend, the detection frequency is dynamically adjusted, nursing resource allocation is optimized, personalized nursing is achieved, and prognosis and management quality of the patients are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring and nursing, and in particular to a nursing method and system based on a hierarchical monitoring mode. Background Art

[0002] Nursing based on the hierarchical monitoring model is a nursing method that manages patients in layers according to the severity of their condition, nursing needs, and health risks. Its core is to scientifically assess the patient's condition, divide it into different levels of monitoring (such as basic care, key care, critical care, etc.), and allocate corresponding nursing resources and intervention measures for each level. Nursing staff dynamically monitor patients according to the grading standards, adjust nursing plans in a timely manner, ensure the rational allocation of resources, and meet the personalized needs of patients to the greatest extent possible to optimize the treatment effect. This model emphasizes precision nursing, improves the quality of nursing through efficient management, reduces the incidence of complications and medical risks, and ultimately improves the patient's overall prognosis and medical experience.

[0003] The prior art has the following deficiencies:

[0004] In the prior art, critical care patients with severe burns usually use a fixed detection cycle to evaluate changes in skin microcirculation in order to establish a regular monitoring mechanism to evaluate tissue perfusion and wound recovery. However, when the patient has local microcirculatory disorders due to the formation of small blood clots in the hypercoagulable state of the blood, the fixed detection cycle cannot adapt to the dynamic changes of the patient's condition, which may lead to serious consequences. Small blood clots hinder the blood supply in the microcirculation, causing the tissue to be in a state of ischemia and hypoxia for a long time, unable to obtain sufficient nutrition and oxygen support, and eventually causing large-scale necrosis of local tissues. Necrotic tissue not only hinders the healing of burn wounds, but may also further extend to surrounding healthy tissues, significantly increasing the risk of infection. This deterioration will cause a significant expansion of the wound area, causing irreparable damage to the patient's recovery.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a nursing method and system based on a hierarchical monitoring mode. By real-time monitoring of skin microcirculation changes and combining intelligent prediction of deep learning models, the scheme accurately captures the early characteristics of blood hypercoagulability trends, such as changes in microcirculatory shear stress and pulsation intensity, generates a blood flow index and compares it with a reference threshold, identifies hypercoagulability early and intervenes in time, and significantly reduces the risk of complications such as local tissue necrosis, spread of infection, and sepsis. Based on intelligent division, patients are divided into two categories: "blood rheology balance" and "blood hypercoagulability trend", the detection frequency is dynamically adjusted, the allocation of nursing resources is optimized, personalized nursing is achieved, and patient prognosis is effectively improved. Improve the management effect of critically ill burn patients to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a nursing method based on a hierarchical monitoring mode, comprising the following steps:

[0008] First, the skin microcirculation perfusion status of severely burned patients is regularly tested according to the preset detection cycle to evaluate tissue perfusion and wound recovery;

[0009] During the test, the patient's skin microcirculation changes are acquired in real time, and subsequent analysis provides continuous and comprehensive raw data;

[0010] Preprocess the acquired data, and based on the preprocessed data, select key features reflecting the trend of the patient's blood in the blood vessels toward hypercoagulation, and perform detailed analysis on the extracted key features within the detection window to quantify the trend of the patient's blood changes;

[0011] The key features that have undergone quantitative analysis are input into the pre-trained deep learning model, and the deep learning model is used to make intelligent predictions on the patient's blood change trends;

[0012] According to the prediction results of the deep learning model, the patient's blood change status is divided into two categories: "blood hypercoagulability tendency" and "blood rheological balance";

[0013] For patients in a state of hemorheological balance, there is no need to increase the frequency of testing, and routine care and monitoring should be carried out according to the preset cycle;

[0014] For patients in a state of hypercoagulable blood tendency, by dynamically increasing the frequency of skin microcirculation change detection, further microvascular obstruction characteristics can be captured in time to prevent the patient's condition from worsening.

[0015] Preferably, based on the preprocessed data, key features reflecting the development of the patient's intravascular blood towards a hypercoagulable trend are screened out, and the screened features include the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow. Under the detection window, the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow are analyzed to generate a microcirculation shear stress reference value and a microcirculation pulsation intensity reference value, respectively, and the trend of the patient's blood changes is quantified by the microcirculation shear stress reference value and the microcirculation pulsation intensity reference value.

[0016] Preferably, the specific steps of analyzing the shear stress exerted on the blood vessel wall when the blood flows in the microvessels under the detection window to generate the microcirculation shear stress reference value are as follows:

[0017] In the detection window, a laser Doppler instrument is used to collect data on blood flow velocity and blood viscosity in microvessels in real time. The formula is as follows:

[0018]

[0019] Where v(x, y) is the blood flow velocity in the microvessel, which indicates the blood flow velocity at the coordinate point (x, y) in the detection area; P(x) and P(y) are pressure gradients, which respectively indicate the pressure change rate in the x-axis direction and the y-axis direction when the blood flows in the microvessel. is the rate of change of pressure along the x-axis, is the rate of change of pressure along the y-axis, and are used to reflect the local pressure difference, η(x, y) is the blood viscosity, K is the fluidity parameter, β is the shear thinning coefficient, It is the rate of change of blood flow velocity in space, reflecting the intensity of local blood flow;

[0020] According to the blood flow velocity v(x, y) and blood viscosity η(x, y) in the microvessels, the shear stress at each point on the microvessel wall is calculated. The calculation expression is as follows:

[0021] ,

[0022] Where τ(x, y) is the microvascular wall shear stress, which describes the tangential force exerted by blood on the vascular wall. It is the velocity gradient along the normal direction n, which reflects the impact of blood flow on the wall. n is the normal direction, which means the direction perpendicular to the blood vessel wall and is used to define the calculation direction of the blood flow velocity gradient.

[0023] The shear stress τ(x, y) of all microvascular walls in the detection area is integrated and calculated to generate the total shear stress of the area. The generation formula is as follows:

[0024] Γ=∫∫ Ω τ(x, y)·ω(x, y)dx dy

[0025] , where Γ is the total shear stress in the detection area, ω(x, y) is the weight function, and Ω is the detection area range, which refers to the two-dimensional plane range of the detection area;

[0026] According to the total shear stress Γ, combined with the nonlinear conversion function, the microcirculatory shear stress reference value is generated, and the generation formula is as follows:

[0027] ,

[0028] In the formula, τ ref is the microcirculatory shear stress reference value, α is the proportional adjustment parameter, λ is the threshold parameter, and γ is the nonlinear sensitivity parameter.

[0029] Preferably, the specific steps of analyzing the intensity of the pulsation signal in the blood flow under the detection window to generate the microcirculation pulsation intensity reference value are as follows:

[0030] First, the spectrum information of the blood flow pulsation signal is collected by laser Doppler equipment, and the amplitude and period characteristics are extracted. The formula is as follows:

[0031] ,

[0032] In the formula, t i is the starting time of the ith pulse signal, Δ is the time interval, S v (v) is the frequency spectrum function of the blood flow velocity signal, v is the frequency variable, L is the upper limit of the frequency range, is the weight adjustment factor, π is the periodicity adjustment factor, which is used to define the periodicity of the cosine function, P(t i ) is the intensity characteristic value of the microcirculation pulsation signal in the i-th segment;

[0033] For the amplitude of the pulsating signal, its energy density weight is calculated to capture the contribution of different frequency components and analyze the exponential decay characteristics of the high-frequency part. The formula is as follows:

[0034] ,

[0035] Where W i is the energy density weight of the ith pulse signal, A k is the amplitude value of the pulsating signal corresponding to the frequency point k, e -ωk is the frequency attenuation factor, e is the natural base, ω is the attenuation coefficient, R k is the signal strength distribution function, N is the total number of frequency points in the detection window, is the normalization coefficient, is the exponential decay coefficient;

[0036] Further analyze the regularity and symmetry of the pulsation signal and calculate the asymmetric characteristics of the signal to capture the distribution anomaly of the pulsation signal under the high coagulation trend. The calculation expression is as follows:

[0037] ,

[0038] In the formula, A i is the asymmetric eigenvalue of the ith pulsating signal, is a nonlinear regulation function;

[0039] The comprehensive pulse signal strength characteristic value P(t i ), the energy density weight W of the pulsating signal i And the asymmetric eigenvalue A i , generate the reference value of microcirculation pulsation intensity, the generation formula is as follows:

[0040] ,

[0041] Where MPI is the reference value of microcirculatory pulsation intensity, θ is the standardized adjustment coefficient, and M is the number of signal segments in the detection window.

[0042] Preferably, after analyzing the screened key features, the generated microcirculation shear stress reference value and microcirculation pulsation intensity reference value are input into a pre-trained deep learning model, and a blood flow index is generated by the deep learning model, and the blood flow index is used to intelligently predict the patient's blood change trend.

[0043] Preferably, the blood flow index generated when the pre-trained deep learning model is used to intelligently predict the patient's blood change trend is compared and analyzed with a pre-set blood flow index reference threshold, and the patient's blood change state is divided, and the division steps are as follows:

[0044] If the blood flow index is greater than or equal to a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood rheological balance";

[0045] If the blood flow index is less than a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood hypercoagulability trend".

[0046] Preferably, for patients in a state of blood hypercoagulation tendency, the specific steps of dynamically increasing the frequency of skin microcirculation change detection to timely capture further microvascular obstruction characteristics are as follows:

[0047] If the patient is in a "hypercoagulable trend", the detection cycle is dynamically adjusted according to the degree of deviation between the current blood flow index BFDI value and the blood flow index reference threshold. The formula is as follows:

[0048] ,

[0049] Where, T adjusted is the adjusted detection period, T preset is the preset detection cycle, BFDI is the blood flow index, BFDI ref is the blood flow index reference threshold, h is the adjustment coefficient;

[0050] Based on the adjusted detection cycle T adjusted , implement a new dynamic testing plan for patients, record the continuous changes in blood flow index BFDI and analyze the trend in real time, update the test data after each test, and calculate the patient's status trend indicator formula as follows:

[0051] ,

[0052] In the formula, STI is the state trend index, which is used to quantify the speed of change of the patient's blood state, and ΔBFDI is the change in the blood flow index BFDI between two consecutive tests;

[0053] According to the STI value of the status trend indicator detected multiple times in succession, the detection cycle and monitoring strategy are iteratively adjusted. If the patient's BFDI begins to approach BFDI ref , then gradually extend the detection cycle; otherwise, further shorten the cycle and strengthen intervention. The formula for updating the detection cycle is as follows:

[0054] ,

[0055] In the formula, ΔSTI is the change of the latest trend indicator relative to the basic trend indicator, STI baseline is the reference trend indicator of the initial state, G is the sensitivity coefficient, T updated is the updated detection period.

[0056] A nursing system based on a hierarchical monitoring mode includes a regular detection module, a data acquisition module, a feature extraction module, an intelligent prediction module, a state division module, a routine monitoring module and a dynamic adjustment module:

[0057] The regular detection module, firstly, regularly detects the skin microcirculation perfusion status of severely burned patients according to the preset detection cycle to evaluate the tissue perfusion and wound recovery;

[0058] The data acquisition module acquires the patient's skin microcirculation changes in real time during the test, and provides continuous and comprehensive raw data for subsequent analysis;

[0059] The feature extraction module pre-processes the acquired data, and based on the pre-processed data, selects key features that reflect the trend of the patient's blood in the blood vessels toward hypercoagulation, and analyzes the extracted key features in detail within the detection window to quantify the trend of the patient's blood changes;

[0060] The intelligent prediction module inputs the key features that have undergone quantitative analysis into a pre-trained deep learning model, and uses the deep learning model to make intelligent predictions on the patient's blood change trends;

[0061] The state classification module divides the patient's blood change state into two categories: "blood hypercoagulability tendency" and "blood rheological balance" according to the prediction results of the deep learning model;

[0062] Routine monitoring module: For patients in a state of blood rheological balance, there is no need to increase the detection frequency, and routine care and monitoring can be carried out according to the preset cycle;

[0063] The dynamic adjustment module dynamically increases the frequency of skin microcirculation change detection for patients in a state of blood hypercoagulation tendency, timely captures further microvascular obstruction characteristics, and prevents the patient's condition from worsening.

[0064] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0065] The present invention monitors the changes in skin microcirculation in real time and combines it with the intelligent prediction of the deep learning model. The scheme can timely capture and quantify the early characteristics of blood hypercoagulation tendency. Key features such as changes in microcirculatory shear stress and pulsation intensity, as well as the generated microcirculatory shear stress reference value and pulsation intensity reference value, provide a reliable basis for the calculation of the blood flow index. By comparing the blood flow index with the reference threshold, it is possible to accurately determine whether the patient has a blood hypercoagulation tendency, so as to take targeted intervention measures in the early stages of tissue ischemia, hypoxia and small thrombus formation. Compared with the traditional fixed detection cycle method, this monitoring mode with dynamic frequency adjustment significantly reduces the risk of serious complications such as local tissue necrosis, spread of infection and sepsis, and prevents further deterioration of the disease. Through early intervention, the patient's rehabilitation quality has been significantly improved, avoiding irreversible tissue damage and life-threatening dangers.

[0066] The present invention uses a deep learning model to intelligently divide the patient's blood change status, and divides the patients into two categories: "blood rheology balance" and "blood hypercoagulability trend", accurately matching nursing resources and testing frequencies. For patients in the state of blood rheology balance, routine monitoring continues to be carried out using the preset detection cycle to reduce unnecessary frequent testing and avoid waste of medical resources; while for patients in the state of blood hypercoagulability trend, the monitoring frequency is dynamically increased, and further changes in microcirculation obstruction are closely monitored to ensure that high-risk patients receive timely intervention. This hierarchical monitoring model not only optimizes the nursing process and improves the utilization efficiency of nursing resources, but also realizes highly personalized monitoring plans to ensure that patients with different conditions can receive targeted nursing support. Accurate nursing strategies significantly improve patient prognosis, and at the same time provide a scientific and efficient solution for the management of critically ill burn patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0068] Figure 1 The present invention is a method flow chart of a nursing method based on a hierarchical monitoring model.

[0069] Figure 2 The module diagram of a nursing system based on a hierarchical monitoring mode of the present invention. DETAILED DESCRIPTION

[0070] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0071] The present invention provides Figure 1 A nursing method based on a hierarchical monitoring model is shown, comprising the following steps:

[0072] First, the skin microcirculation perfusion status of severely burned patients is regularly tested according to the preset testing cycle (e.g., every hour, every four hours, etc.) to evaluate tissue perfusion and wound recovery.

[0073] The main purpose of this step is to establish a basic monitoring framework to ensure that all patients are systematically monitored in the initial stage. Through fixed-cycle testing, basic data on the patient's skin microcirculation can be obtained in a timely manner to assess the initial situation of tissue perfusion and wound recovery. This regular monitoring mechanism helps to standardize the nursing process, ensure that no patients who need attention are missed, and provide baseline data for subsequent dynamic adjustments.

[0074] During the test, the patient's skin microcirculation changes are acquired in real time, and subsequent analysis provides continuous and comprehensive raw data;

[0075] Advanced monitoring equipment (such as laser Doppler, infrared thermal imager, etc.) is used to obtain real-time data information on changes in the patient's skin microcirculation. These devices can continuously and accurately record key indicators such as blood flow velocity, blood flow, and perfusion reference values ​​in the microvessels. The purpose of real-time data acquisition is to ensure that nursing staff can grasp the dynamic changes of the patient's microcirculation in a timely manner, capture potential abnormal signals, and provide a rich source of information for subsequent data analysis and feature extraction.

[0076] Preprocess the acquired data, and based on the preprocessed data, select key features reflecting the trend of the patient's blood in the blood vessels toward hypercoagulation, and perform detailed analysis on the extracted key features within the detection window to quantify the trend of the patient's blood changes;

[0077] The acquired real-time data often contains noise, outliers or incomplete information, so preprocessing is required. Preprocessing steps include data cleaning, filtering, normalization, etc. to improve the quality and availability of data. Through preprocessing, irrelevant or interfering information can be eliminated to ensure that subsequent feature extraction and analysis are more accurate and reliable. Preprocessed data will be more suitable for input into deep learning models, improving the prediction accuracy and stability of the model.

[0078] Based on the preprocessed data, key features reflecting the development of the hypercoagulable trend of the blood in the patient's blood vessels are screened out, and the screened features include the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow. Under the detection window, the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow are analyzed to generate a microcirculation shear stress reference value and a microcirculation pulsation intensity reference value respectively, and the trend of the patient's blood change is quantified through the microcirculation shear stress reference value and the microcirculation pulsation intensity reference value;

[0079] The shear stress exerted on the blood vessel wall by blood flowing in the microvessels continues to decrease, which clearly indicates that the patient's blood is currently developing towards a hypercoagulable trend. This is because the change in shear stress directly reflects the change in the rheological properties of the blood. When the shear stress decreases, it usually means that the blood flow rate slows down and the blood viscosity increases, and these changes are the early manifestations of the hypercoagulable state. Continuous low shear stress can cause damage to the function of vascular endothelial cells, weaken their anticoagulant and anti-inflammatory abilities, thereby promoting enhanced platelet adhesion, accelerated aggregation, and activating the coagulation factor cascade reaction, further aggravating the hypercoagulable trend of the blood. In addition, the low shear stress environment provides ideal conditions for the formation of small thrombi, which may lead to insufficient microcirculatory perfusion and local tissue ischemia and hypoxia. If the shear stress continues to decrease and is not intervened in time, it may cause microcirculatory obstruction, leading to local tissue necrosis and other complications. Therefore, the continuous decrease in shear stress is not only a hallmark of the hypercoagulable state, but also indicates that the patient's blood is rapidly developing towards a more dangerous hypercoagulable trend, and measures should be taken immediately for intervention and treatment.

[0080] The specific steps for analyzing the shear stress exerted on the blood vessel wall when the blood flows in the microvessels under the detection window to generate the microcirculation shear stress reference value are as follows:

[0081] In the detection window, a laser Doppler instrument is used to collect data on blood flow velocity and blood viscosity in microvessels in real time. The data is based on two-dimensional spatial coordinates and covers the entire detection area. The formula is as follows:

[0082] Where v(x, y) is the blood flow velocity in the microvessel, which indicates the blood flow velocity at the coordinate point (x, y) in the detection area; P(x) and P(y) are pressure gradients, which respectively indicate the pressure change rate in the x-axis direction and the y-axis direction when the blood flows in the microvessel. is the rate of change of pressure along the x-axis, is the rate of change of pressure along the y-axis, and Both are used to reflect local pressure differences. η(x, y) is blood viscosity. K is a fluidity parameter that characterizes blood fluidity and is affected by blood components (such as the number of red blood cells and fibrinogen concentration). β is the shear thinning coefficient, which reflects the sensitivity of blood viscosity to changes in flow gradient. It is used to adjust the dependence of blood viscosity on flow gradient. The larger the value, the more sensitive the viscosity is to changes in flow velocity. It is the rate of change of blood flow velocity in space, reflecting the intensity of local blood flow;

[0083] This step dynamically describes the distribution of blood flow velocity and viscosity in the detection area, providing a basis for the subsequent calculation of shear stress.

[0084] According to the blood flow velocity v(x, y) and blood viscosity η(x, y) in the microvessels, the shear stress at each point on the microvessel wall is calculated. The calculation expression is as follows:

[0085] ,

[0086] Where τ(x, y) is the microvascular wall shear stress, which describes the tangential force exerted by blood on the vascular wall. It is the velocity gradient along the normal direction n, which reflects the impact of blood flow on the wall. n is the normal direction, which means the direction perpendicular to the blood vessel wall and is used to define the calculation direction of the blood flow velocity gradient.

[0087] This step combines blood viscosity and flow velocity gradient to generate a shear stress distribution map for each coordinate point, which is used to quantitatively analyze stress abnormalities in local areas.

[0088] The shear stress τ(x, y) of all microvascular walls in the detection area is integrated to generate the total shear stress of the area. The generation formula is as follows:

[0089] Γ=∫∫ Ω τ(x, y)·ω(x, y)dx dy

[0090] , where Γ is the total shear stress in the detection area, which comprehensively reflects the overall effect of blood flow in the entire area on the vascular wall, ω(x, y) is the weight function, which is used to weight the local shear stress and indicates the contribution of the shear stress at each coordinate point to the total shear stress Γ, and Ω is the detection area range, which refers to the two-dimensional plane range of the detection area, defined by a laser Doppler instrument or other detection equipment;

[0091] The local shear stress distribution is transformed into a global parameter by integration, highlighting the influence of high stress areas and providing a global perspective for the generation of reference values.

[0092] According to the total shear stress Γ, combined with the nonlinear conversion function, the microcirculatory shear stress reference value is generated, and the generation formula is as follows:

[0093] ,

[0094] In the formula, τ ref is the microcirculatory shear stress reference value, α is the proportional adjustment parameter, λ is the threshold parameter, which is used as the reference shear stress value to normalize the distribution range of Γ, and γ is the nonlinear sensitivity parameter, which controls the response degree to the shear stress change in the formula and determines τ ref Nonlinear amplification effect on Γ.

[0095] Through nonlinear transformation, the differences in high shear stress are amplified and the reference values ​​are standardized to facilitate comparative analysis between different detection windows.

[0096] The proportional adjustment parameter α refers to the parameter used to adjust the overall magnitude of the calculation result to ensure that the value output by the formula is adapted to a specific application range or standard reference range. In the formula for generating the microcirculatory shear stress reference value, the main function of the proportional adjustment parameter is to adjust the reference value by linear scaling to make it consistent and comparable between different detection devices, detection environments or individual patients. For example, different monitoring devices may cause deviations in shear stress values ​​due to different sensitivities, and the proportional adjustment parameter can correct such deviations. In addition, it can also control the magnitude of the calculation result within a range that is easy to interpret and analyze according to specific needs, thereby improving the practicality and explanatory power of the results, thereby making the generated microcirculatory shear stress reference value more stable and reliable.

[0097] The smaller the microcirculation shear stress reference value generated by analyzing the shear stress exerted on the blood vessel wall when the blood flows in the microvessels under the detection window, the more it indicates that the patient's blood is developing towards a hypercoagulable trend. This is because the decrease in the shear stress value reflects the dynamic process of slowing blood flow and increasing blood viscosity, which are the early signs of a hypercoagulable state. Continuous low shear stress will not only weaken the anticoagulant function of vascular endothelial cells, but may also accelerate platelet adhesion and aggregation, activate coagulation factors, and thus lead to the formation of small blood clots and microcirculation obstruction. On the contrary, if the shear stress reference value is maintained within the normal range or even higher, it means that the blood fluidity is good, the blood rheological properties are stable, and the patient's blood is currently in a state of rheological equilibrium and is not developing towards a hypercoagulable trend.

[0098] The gradual weakening of the intensity of the pulsation signal in the blood flow does indicate that the patient's blood is developing towards a hypercoagulable trend, because the intensity of the pulsation signal directly reflects the patency and periodic changes of blood flow in the microcirculation. Under normal circumstances, the intensity of the pulsation signal in the blood flow is driven by the heart pumping blood and transmitted to the microcirculation network through the elastic vascular wall. However, when the patient's blood enters a hypercoagulable trend, the increase in blood viscosity and the enhancement of platelet aggregation will lead to an increase in microvascular resistance, local blood flow obstruction, and decreased blood fluidity, which will gradually weaken or even interrupt the pulsation signal. This change not only indicates that blood flow is restricted, but may also indicate that small thrombi have already formed in the microcirculation, resulting in abnormal blood rheological properties. These hallmark changes in the hypercoagulable state will further aggravate the ischemia and hypoxia of local tissues, and may cause more serious perfusion disorders and tissue damage. Therefore, the weakening of the pulsation signal intensity is an important manifestation of the hypercoagulable trend of blood, providing a key monitoring basis for early identification and intervention of hypercoagulable states.

[0099] The specific steps for analyzing the intensity of the pulsation signal in the blood flow under the detection window to generate the microcirculation pulsation intensity reference value are as follows:

[0100] First, the spectrum information of the blood flow pulsation signal is collected by laser Doppler equipment, and the amplitude and period characteristics are extracted. The goal of this process is to identify the basic dynamic characteristics of blood flow in microcirculation. The formula is as follows:

[0101] ,

[0102] In the formula, t i is the starting time of the ith pulse signal, Δ is the time interval, S v (ν) is the frequency spectrum function of the blood flow velocity signal, v is the frequency variable, L is the upper limit of the frequency range, is the weight adjustment factor, which represents the weighting factor for different frequency components. Its amplitude fluctuates periodically with frequency. π is the periodic adjustment factor, which is used to define the periodicity of the cosine function. i ) is the intensity characteristic value of the microcirculation pulsation signal in the i-th segment;

[0103] This step extracts the important features of the blood flow pulsation signal by integrating and calculating the amplitude and periodicity characteristics of the spectrum signal, providing a basis for subsequent analysis.

[0104] For the amplitude of the pulsating signal, its energy density weight is calculated to capture the contribution of different frequency components and analyze the exponential decay characteristics of the high-frequency part. The formula is as follows:

[0105] ,

[0106] Where W i is the energy density weight of the ith segment of the pulsation signal, indicating the comprehensive strength of the pulsation signal within the detection window. It is an indicator obtained by weighted calculation of the amplitude and energy distribution of different frequency components. It measures the overall intensity characteristics of the entire pulsation signal within the detection window and can reflect the energy distribution of blood flow pulsation. k is the amplitude value of the pulsating signal corresponding to the frequency point k, reflecting the dynamic changes of local blood flow, e -ωk is the frequency attenuation factor, which represents the exponential attenuation weight of the frequency component, e is the natural base, ω is the attenuation coefficient, R k is the signal strength distribution function, which represents the signal strength distribution function corresponding to the kth frequency point, reflecting the energy distribution ratio of the pulsating signal at different frequencies. N is the total number of frequency points in the detection window. It is the normalization coefficient, which is used to normalize all frequency components to ensure the consistency of weight calculation in different frequency ranges and detection windows. Through exponential attenuation adjustment, the signal in the high-frequency range can be properly smoothed to avoid excessive amplification of the influence of high-frequency noise. is the exponential decay coefficient;

[0107] This step is achieved by combining the signal amplitude value Ak And the signal strength distribution function R k , calculate the comprehensive energy characteristics of the signal, and use the exponential decay factor to control the influence of the high-frequency part, so as to extract the energy density weight value W i The energy density weight can reflect the overall intensity and distribution characteristics of blood flow pulsation and is one of the important parameters for identifying hypercoagulable trends.

[0108] Further analyze the regularity and symmetry of the pulsation signal and calculate the asymmetric characteristics of the signal to capture the distribution anomaly of the pulsation signal under the high coagulation trend. The calculation expression is as follows:

[0109] ,

[0110] In the formula, A i is the asymmetric characteristic value of the ith segment of the pulsating signal, which is used to quantitatively describe the regularity and symmetry of the pulsating signal in the time domain distribution. is a nonlinear regulation function;

[0111] This step captures the irregularity or offset characteristics of the signal in the distribution through nonlinear transformation of the pulsating signal, thereby generating an asymmetric eigenvalue A. i The hypercoagulability trend is usually manifested as an abnormal deviation or imbalance in the distribution of the pulsation signal. Asymmetry analysis can intuitively reflect these abnormal states and provide a basis for further prediction of the hypercoagulability trend of blood.

[0112] The nonlinear adjustment function is used to perform nonlinear transformation on the relationship between input variables, thereby amplifying or suppressing certain specific features and enhancing the sensitivity of the formula to abnormal changes. The function is to dynamically adjust the pulsation signal strength characteristic value P(t i ) and energy density weight W i It introduces the nonlinear characteristics of the sine function to transform the originally smooth or linear distribution into a more sensitive nonlinear output, thereby capturing slight deviations or asymmetries in the signal. This function is particularly suitable for detecting abnormal signals in complex systems, such as abnormal changes in blood flow pulsation signals under hypercoagulable trends, making the formula more sensitive to features that deviate from the normal state and more accurately identifying early signs of hypercoagulable trends.

[0113] The comprehensive pulse signal strength characteristic value P(t i ), the energy density weight W of the pulsating signal i And the asymmetric eigenvalue A i , generate the reference value of microcirculation pulsation intensity, the generation formula is as follows:

[0114] ,

[0115] Where MPI is the reference value of microcirculation pulsation intensity, θ is the standardized adjustment coefficient, which is used to adjust the scale of the calculation result of the formula so that the range of the reference value of microcirculation pulsation intensity can adapt to data analysis in different equipment and environments, and M is the number of signal segments in the detection window, indicating the number of segmentations when performing signal analysis in the window.

[0116] This step integrates the asymmetry, energy weight and intensity characteristics of the pulsation signal, and generates the microcirculation pulsation intensity reference value MPI through standardization and square operation. The smaller the reference value, the weaker the blood flow pulsation signal and the worse the regularity, which may indicate that the blood is developing towards a hypercoagulable trend; a higher reference value indicates that the blood is in a balanced rheological state. This value provides a scientific basis for evaluating the patient's microcirculation status and hypercoagulable risk.

[0117] The smaller the performance value of the microcirculation pulsation intensity reference value generated after analyzing the intensity of the pulsation signal in the blood flow under the detection window, the more it indicates that the patient's blood is developing towards a hypercoagulable trend, while a higher performance value indicates that the patient's blood is not developing towards a hypercoagulable trend. This is because the microcirculation pulsation intensity reference value reflects the intensity and regularity of the pulsation signal in the blood flow, and can intuitively reflect the blood flow conditions in the microcirculation. When the performance value is small or decreases, it usually means that the blood flow in the microvessels slows down or the pulsation is obstructed, which is often associated with hypercoagulable trends such as increased blood viscosity and small thrombus formation. Under hypercoagulable conditions, the irregularity of microcirculation blood flow and the increase in resistance will weaken the pulsation signal intensity, resulting in a lower reference value. When the performance value is high and stable, it means that the blood flow is smooth, the blood viscosity is normal, and there is no significant risk of thrombosis, indicating that the patient's blood state is in rheological balance.

[0118] The key features that have undergone quantitative analysis are input into the pre-trained deep learning model, and the deep learning model is used to make intelligent predictions on the patient's blood change trends;

[0119] After analyzing the screened key features, the generated microcirculation shear stress reference value and microcirculation pulsation intensity reference value are input into the pre-trained deep learning model. The blood flow index is generated by the deep learning model, and the blood change trend of the patient is intelligently predicted by the blood flow index.

[0120] A pre-trained deep learning model refers to a model that has been trained and parameter optimized using a large amount of high-quality historical data. It can automatically extract deep features from the input data and make predictions or classifications. During the training phase, the model gradually adjusts its internal weights and bias parameters by learning the mapping relationship between input data (such as microcirculatory shear stress reference values ​​and microcirculatory pulsation intensity reference values) and target results (such as blood flow index and blood change trend), so that it has the ability to accurately fit the nonlinear relationship of input features. In practical applications, the trained model does not need to be built from scratch, but can be directly deployed and applied to the intelligent analysis of real-time data. The advantage of this pre-trained model is that it can efficiently process complex and multi-dimensional medical data, explore hidden patterns and trends, and thus provide accurate predictions for the early identification of hypercoagulable trends.

[0121] The training phase usually requires a large amount of historical case data, including the patient's microcirculatory shear stress, pulsation intensity, and actual blood flow status annotations. These data are input into the model through a specific training algorithm (such as supervised learning), and the model gradually adjusts the parameters according to the predefined loss function until the deviation between its prediction results and the labeled data is minimized. In addition, the model may use technologies such as convolutional neural networks (CNN) or recurrent neural networks (RNN) to process complex time series data or spatial features. In this way, the deep learning model can capture the subtle relationship between microcirculatory characteristics and blood flow status, thereby having the ability to make intelligent predictions.

[0122] In this scheme, the pre-trained deep learning model further generates a blood flow index by inputting the microcirculation shear stress reference value and the pulsation intensity reference value, and intelligently predicts the patient's blood change trend based on the index. The realization of this predictive ability depends on the knowledge accumulated by the model during the training phase. By learning the historical data of the patient population, the model identifies those patterns in the microcirculation characteristics that are closely related to the hypercoagulable trend, such as abnormally reduced shear stress and significantly weakened pulsation signals. In addition, the model can also comprehensively consider the interactive relationship between multidimensional features and identify nonlinear change trends in complex pathological conditions. For example, when the shear stress and pulsation intensity are both abnormal, the model can accurately predict the abnormal changes in the blood flow index, thereby suggesting that the patient may be developing a hypercoagulable trend.

[0123] The intelligent prediction capability of the pre-trained model significantly improves the efficiency and accuracy of blood rheology trend analysis. Compared with traditional statistical analysis methods, the deep learning model has stronger generalization and adaptability. This intelligent trend prediction can not only detect potential hypercoagulable risks in a timely manner, but also provide clear decision-making support information for nursing staff, such as dynamically adjusting the monitoring cycle or initiating intervention measures. By utilizing the pre-trained deep learning model, this solution can convert complex microcirculation characteristic data into intuitive blood flow indexes, greatly improving the intelligence and precision of patient condition assessment.

[0124] The deep learning model is not limited here and can convert the microcirculatory shear stress reference value τ ref A deep learning model for generating a blood flow index BFDI by comprehensive analysis with a microcirculation pulsation intensity reference value MPI can be used. To implement the technical solution of the present invention, the present invention provides a specific generation method;

[0125] The blood flow index BFDI is generated by the following formula:

[0126] ,

[0127] Where Q1 and Q2 are the microcirculatory shear stress reference values ​​τ ref And the preset proportional coefficient of the microcirculation pulsation intensity reference value MPI, and Q1 and Q2 are both greater than 0.

[0128] It can be seen from the blood flow index that the smaller the microcirculation shear stress reference value generated after analyzing the shear stress exerted on the blood vessel wall when the blood flows in the microvessels under the detection window, the smaller the microcirculation pulsation intensity reference value generated after analyzing the intensity of the pulsation signal in the blood flow under the detection window, indicating that the smaller the blood flow index expression value generated when the patient's blood change trend is intelligently predicted by the pre-trained deep learning model, the greater the probability that the patient's blood is developing towards a hypercoagulable trend, and vice versa, the smaller the probability that the patient's blood is developing towards a hypercoagulable trend.

[0129] According to the prediction results of the deep learning model, the patient's blood change status is divided into two categories: "blood hypercoagulability tendency" and "blood rheological balance";

[0130] The blood flow index generated by the pre-trained deep learning model when intelligently predicting the patient's blood change trend is compared and analyzed with the pre-set blood flow index reference threshold, and the patient's blood change status is divided. The division steps are as follows:

[0131] If the blood flow index is greater than or equal to a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood rheological balance";

[0132] If the blood flow index is less than a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood hypercoagulability trend".

[0133] "Blood hypercoagulability" refers to the process in which the patient's blood gradually develops from normal rheological properties to a hypercoagulable state, which is manifested by an overactive blood coagulation system, enhanced platelet aggregation, increased blood viscosity, and the possible formation of small blood clots in the microcirculation. "Blood rheological balance" means that the patient's blood is in normal rheological properties, the coagulation, anticoagulation and fibrinolytic systems are in a dynamic equilibrium state, the blood flows smoothly and evenly in the microcirculation, and there is no obvious hypercoagulability or hypocoagulability tendency.

[0134] For patients in a state of hemorheological balance, there is no need to increase the frequency of testing, and routine care and monitoring should be carried out according to the preset cycle;

[0135] For patients in a state of hemorheological balance, routine care and monitoring are carried out according to the preset cycle without increasing the frequency of testing. The purpose is to optimize the allocation of nursing resources while avoiding unnecessary interference. The state of hemorheological balance indicates that the patient's blood fluidity and microcirculation function are normal and the risk is low. Frequent testing can neither significantly improve the sensitivity of problem identification nor increase patient discomfort or waste of resources. By following a fixed cycle for routine monitoring, it can be ensured that patients receive sufficient attention and basic care, while human resources, equipment and other resources are concentrated on high-risk patients to improve the overall nursing efficiency and quality, thereby providing more scientific and reasonable resource management support for the medical team.

[0136] For patients with a high blood coagulation tendency, the frequency of skin microcirculation change detection can be dynamically increased to timely capture further microvascular obstruction characteristics and prevent the patient's condition from worsening;

[0137] For patients in a state of hypercoagulable blood, the specific steps to dynamically increase the frequency of skin microcirculation change detection and timely capture further microvascular obstruction characteristics are as follows:

[0138] If the patient is in a "hypercoagulable trend", the detection cycle is dynamically adjusted according to the degree of deviation between the current blood flow index BFDI value and the blood flow index reference threshold. The formula is as follows:

[0139] ,

[0140] Where, T adjusted is the adjusted detection period, T preset is the preset detection cycle, which refers to the fixed detection frequency under normal circumstances. BFDI is the blood flow index. refis the blood flow index reference threshold, h is the adjustment coefficient, which is used to balance the adjustment amplitude of the detection cycle;

[0141] When the blood flow index BFDI is significantly lower than the blood flow index reference threshold BFDI ref When the patient's microcirculation status is monitored more frequently, the adjusted detection cycle will be greatly shortened.

[0142] Based on the adjusted detection cycle T adjusted , implement a new dynamic testing plan for patients, record the continuous changes in blood flow index BFDI and analyze the trend in real time, update the test data after each test, and calculate the patient's status trend indicator formula as follows:

[0143] ,

[0144] In the formula, STI is the state trend index, which is used to quantify the speed of change of the patient's blood state, and ΔBFDI is the change in the blood flow index BFDI between two consecutive tests, reflecting the improvement or deterioration trend of blood flow;

[0145] If the STI shows a continuous negative value and an increase in the absolute value (i.e., the blood flow index continues to decrease), it indicates that the patient's hypercoagulability trend is aggravated and a higher level of intervention and monitoring is required.

[0146] According to the STI value of the status trend indicator detected multiple times in succession, the detection cycle and monitoring strategy are iteratively adjusted. If the patient's BFDI begins to approach BFDI ref , then gradually extend the detection cycle; otherwise, further shorten the cycle and strengthen intervention. The formula for updating the detection cycle is as follows:

[0147] ,

[0148] In the formula, ΔSTI is the change of the latest trend indicator relative to the basic trend indicator, STI baseline is the reference trend indicator of the initial state, which is used to measure the baseline change of the patient's state. G is the sensitivity coefficient, which is used to smooth the cycle adjustment amplitude. updated It is the updated detection period, which is dynamically adjusted based on the status trend indicator.

[0149] Through cyclic updates, we ensure that the testing frequency always dynamically matches the patient's actual condition, avoiding insufficient testing or waste of resources, while providing real-time basis for clinical decision-making.

[0150] For patients in a state of hypercoagulable blood trend, by dynamically increasing the frequency of skin microcirculation change detection, further microvascular obstruction characteristics can be captured in time. Its core role is to enhance the sensitivity and timeliness of real-time monitoring, thereby providing a scientific basis for early intervention and precise care for patients. When patients are in a state of hypercoagulable blood trend, blood viscosity increases and small blood clots gradually form. This state may rapidly develop into complete obstruction of local microvessels, leading to serious consequences such as local tissue ischemia, hypoxia, necrosis, and expansion of the wound area. Increasing the detection frequency can identify these abnormal microcirculatory changes more quickly, capture characteristics such as reduced blood flow rate, reduced perfusion volume, and reduced shear stress, and help the nursing team take targeted measures as early as possible, such as optimizing anticoagulant therapy, adjusting infusion strategies, or strengthening wound management.

[0151] In addition, the dynamic adjustment of frequency makes the monitoring strategy more consistent with the actual changes in the patient's condition, avoiding the lag problem that may be caused by a fixed detection cycle, so that intervention can be implemented before the condition has deteriorated to an irreversible stage. A higher monitoring frequency can also accumulate more data points, capture subtle trends of abnormal changes, and provide the medical team with a reliable basis for dynamic condition assessment. Through this enhanced monitoring, the risk of complications caused by microcirculation obstruction, such as spread of infection, deterioration of wound surface, and multiple organ dysfunction, can be effectively reduced, thereby improving the patient's prognosis and improving the overall quality of care and resource utilization efficiency. This process not only protects the patient's life safety, but also reduces the blind spots of care in high-risk conditions, making nursing intervention more scientific and effective.

[0152] The present invention monitors the changes in skin microcirculation in real time and combines it with the intelligent prediction of the deep learning model. The scheme can timely capture and quantify the early characteristics of blood hypercoagulation tendency. Key features such as changes in microcirculatory shear stress and pulsation intensity, as well as the generated microcirculatory shear stress reference value and pulsation intensity reference value, provide a reliable basis for the calculation of the blood flow index. By comparing the blood flow index with the reference threshold, it is possible to accurately determine whether the patient has a blood hypercoagulation tendency, so as to take targeted intervention measures in the early stages of tissue ischemia, hypoxia and small thrombus formation. Compared with the traditional fixed detection cycle method, this monitoring mode with dynamic frequency adjustment significantly reduces the risk of serious complications such as local tissue necrosis, spread of infection and sepsis, and prevents further deterioration of the disease. Through early intervention, the patient's rehabilitation quality has been significantly improved, avoiding irreversible tissue damage and life-threatening dangers.

[0153] The present invention uses a deep learning model to intelligently divide the patient's blood change status, and divides the patients into two categories: "blood rheology balance" and "blood hypercoagulability trend", accurately matching nursing resources and testing frequencies. For patients in the state of blood rheology balance, routine monitoring continues to be carried out using the preset detection cycle to reduce unnecessary frequent testing and avoid waste of medical resources; while for patients in the state of blood hypercoagulability trend, the monitoring frequency is dynamically increased, and further changes in microcirculation obstruction are closely monitored to ensure that high-risk patients receive timely intervention. This hierarchical monitoring model not only optimizes the nursing process and improves the utilization efficiency of nursing resources, but also realizes highly personalized monitoring plans to ensure that patients with different conditions can receive targeted nursing support. Accurate nursing strategies significantly improve patient prognosis, and at the same time provide a scientific and efficient solution for the management of critically ill burn patients.

[0154] The present invention provides Figure 2 A nursing system based on a hierarchical monitoring mode is shown, comprising a regular detection module, a data acquisition module, a feature extraction module, an intelligent prediction module, a state division module, a routine monitoring module and a dynamic adjustment module:

[0155] The regular detection module, firstly, regularly detects the skin microcirculation perfusion status of severely burned patients according to the preset detection cycle to evaluate the tissue perfusion and wound recovery;

[0156] The data acquisition module acquires the patient's skin microcirculation changes in real time during the test, and provides continuous and comprehensive raw data for subsequent analysis;

[0157] The feature extraction module pre-processes the acquired data, and based on the pre-processed data, selects key features that reflect the trend of the patient's blood in the blood vessels toward hypercoagulation, and analyzes the extracted key features in detail within the detection window to quantify the trend of the patient's blood changes;

[0158] The intelligent prediction module inputs the key features that have undergone quantitative analysis into a pre-trained deep learning model, and uses the deep learning model to make intelligent predictions on the patient's blood change trends;

[0159] The state classification module divides the patient's blood change state into two categories: "blood hypercoagulability tendency" and "blood rheological balance" according to the prediction results of the deep learning model;

[0160] Routine monitoring module: For patients in a state of blood rheological balance, there is no need to increase the detection frequency, and routine care and monitoring can be carried out according to the preset cycle;

[0161] The dynamic adjustment module dynamically increases the frequency of skin microcirculation change detection for patients in a state of blood hypercoagulation tendency, timely captures further microvascular obstruction characteristics, and prevents the patient's condition from worsening.

[0162] A nursing method based on a hierarchical monitoring model provided in an embodiment of the present invention is implemented through the above-mentioned nursing system based on a hierarchical monitoring model. The specific methods and processes of a nursing system based on a hierarchical monitoring model are detailed in the embodiment of the above-mentioned nursing method based on a hierarchical monitoring model, which will not be repeated here.

[0163] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0164] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0165] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0166] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0167] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0168] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0169] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0171] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0172] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A nursing method based on a hierarchical monitoring model, characterized in that: The following steps are involved: First, the skin microcirculation perfusion status of severely burned patients is regularly tested according to the preset detection cycle to evaluate tissue perfusion and wound recovery; During the test, the patient's skin microcirculation changes are acquired in real time, and subsequent analysis provides continuous and comprehensive raw data; Preprocess the acquired data, and based on the preprocessed data, select key features reflecting the trend of the patient's blood in the blood vessels toward hypercoagulation, and perform detailed analysis on the extracted key features within the detection window to quantify the trend of the patient's blood changes; The key features that have undergone quantitative analysis are input into the pre-trained deep learning model, and the deep learning model is used to make intelligent predictions on the patient's blood change trends; According to the prediction results of the deep learning model, the patient's blood change status is divided into two categories: "blood hypercoagulability trend" and "blood rheological balance"; For patients in a state of hemorheological balance, there is no need to increase the frequency of testing, and routine care and monitoring should be carried out according to the preset cycle; For patients in a state of hypercoagulable blood tendency, by dynamically increasing the frequency of skin microcirculation change detection, further microvascular obstruction characteristics can be captured in time to prevent the patient's condition from worsening.

2. A nursing method based on a hierarchical monitoring model according to claim 1, characterized in that: Based on the preprocessed data, key features reflecting the development of the patient's intravascular blood towards hypercoagulation trend are screened out, and the screened features include the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow. Under the detection window, the shear stress exerted on the blood vessel wall when the blood flows in the microvessels and the intensity of the pulsation signal in the blood flow are analyzed to generate microcirculation shear stress reference values ​​and microcirculation pulsation intensity reference values, respectively. The trend of patient blood changes is quantified by the microcirculation shear stress reference values ​​and microcirculation pulsation intensity reference values.

3. A nursing method based on a hierarchical monitoring model according to claim 2, characterized in that: The specific steps for analyzing the shear stress exerted on the blood vessel wall when the blood flows in the microvessels under the detection window to generate the microcirculation shear stress reference value are as follows: In the detection window, a laser Doppler instrument is used to collect data on blood flow velocity and blood viscosity in microvessels in real time. The formula is as follows: Where v(x, y) is the blood flow velocity in the microvessel, which indicates the blood flow velocity at the coordinate point (x, y) in the detection area; P(x) and P(y) are pressure gradients, which respectively indicate the pressure change rate in the x-axis direction and the y-axis direction when the blood flows in the microvessel. is the rate of change of pressure along the x-axis, is the rate of change of pressure along the y-axis, and are used to reflect the local pressure difference, η(x, y) is the blood viscosity, K is the fluidity parameter, β is the shear thinning coefficient, It is the rate of change of blood flow velocity in space, reflecting the intensity of local blood flow; According to the blood flow velocity v(x, y) and blood viscosity η(x, y) in the microvessels, the shear stress at each point on the microvessel wall is calculated. The calculation expression is as follows: ， Where τ(x, y) is the microvascular wall shear stress, which describes the tangential force exerted by blood on the vascular wall. It is the velocity gradient along the normal direction n, which reflects the impact of blood flow on the wall. n is the normal direction, which means the direction perpendicular to the blood vessel wall and is used to define the calculation direction of the blood flow velocity gradient. The shear stress τ(x, y) of all microvascular walls in the detection area is integrated and calculated to generate the total shear stress of the area. The generation formula is as follows: Γ=∫∫ Ω τ(x,y)·ω(x,y)dx dy, Where Γ is the total shear stress in the detection area, ω(x, y) is the weight function, and Ω is the detection area range, which refers to the two-dimensional plane range of the detection area; According to the total shear stress Γ, combined with the nonlinear conversion function, the microcirculatory shear stress reference value is generated, and the generation formula is as follows: ， In the formula, τ ref is the microcirculatory shear stress reference value, α is the proportional adjustment parameter, λ is the threshold parameter, and γ is the nonlinear sensitivity parameter.

4. A nursing method based on a hierarchical monitoring model according to claim 2, characterized in that: The specific steps for analyzing the intensity of the pulsation signal in the blood flow under the detection window to generate the microcirculation pulsation intensity reference value are as follows: First, the spectrum information of the blood flow pulsation signal is collected by laser Doppler equipment, and the amplitude and period characteristics are extracted. The formula is as follows: ， Where, t i is the starting time of the ith pulse signal, Δ is the time interval, S v (v) is the frequency spectrum function of the blood flow velocity signal, v is the frequency variable, L is the upper limit of the frequency range, is the weight adjustment factor, π is the periodicity adjustment factor, which is used to define the periodicity of the cosine function, P(t i ) is the intensity characteristic value of the microcirculation pulsation signal in the i-th segment; For the amplitude of the pulsating signal, its energy density weight is calculated to capture the contribution of different frequency components and analyze the exponential decay characteristics of the high-frequency part. The formula is as follows: ， Where W i is the energy density weight of the ith pulse signal, A k is the amplitude value of the pulsating signal corresponding to the frequency point k, e -ωk is the frequency attenuation factor, e is the natural base, ω is the attenuation coefficient, R k is the signal strength distribution function, N is the total number of frequency points in the detection window, is the normalization coefficient, is the exponential decay coefficient; Further analyze the regularity and symmetry of the pulsation signal and calculate the asymmetric characteristics of the signal to capture the distribution anomaly of the pulsation signal under the high coagulation trend. The calculation expression is as follows: ， In the formula, A i is the asymmetric eigenvalue of the ith pulsating signal, is a nonlinear regulation function; The comprehensive pulse signal strength characteristic value P(t i ), the energy density weight W of the pulsating signal i and the asymmetric eigenvalue A i , generate the reference value of microcirculation pulsation intensity, the generation formula is as follows: ， Where MPI is the reference value of microcirculatory pulsation intensity, θ is the standardized adjustment coefficient, and M is the number of signal segments in the detection window.

5. A nursing method based on a hierarchical monitoring model according to claim 2, characterized in that: After analyzing the screened key features, the generated microcirculation shear stress reference value and microcirculation pulsation intensity reference value are input into the pre-trained deep learning model. The blood flow index is generated by the deep learning model, and the blood change trend of the patient is intelligently predicted by the blood flow index.

6. A nursing method based on a hierarchical monitoring model according to claim 5, characterized in that: The blood flow index generated by the pre-trained deep learning model when intelligently predicting the patient's blood change trend is compared and analyzed with the pre-set blood flow index reference threshold, and the patient's blood change status is divided. The division steps are as follows: If the blood flow index is greater than or equal to a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood rheological balance"; If the blood flow index is less than a preset blood flow index reference threshold, the current patient's blood change state is classified as "blood hypercoagulability trend".

7. A nursing method based on a hierarchical monitoring model according to claim 6, characterized in that: For patients in a state of hypercoagulable blood, the specific steps to dynamically increase the frequency of skin microcirculation change detection and timely capture further microvascular obstruction characteristics are as follows: If the patient is in a "hypercoagulable trend", the detection cycle is dynamically adjusted according to the degree of deviation between the current blood flow index BFDI value and the blood flow index reference threshold. The formula is as follows: ， Where, T adjusted is the adjusted detection period, T preset is the preset detection cycle, BFDI is the blood flow index, BFDI ref is the blood flow index reference threshold, h is the adjustment coefficient; Based on the adjusted detection cycle T adjusted , implement a new dynamic testing plan for patients, record the continuous changes in blood flow index BFDI and analyze the trend in real time, update the test data after each test, and calculate the patient's status trend indicator formula as follows: ， In the formula, STI is the state trend index, which is used to quantify the speed of change of the patient's blood state, and ΔBFDI is the change in the blood flow index BFDI between two consecutive tests; According to the STI value of the status trend indicator detected multiple times in succession, the detection cycle and monitoring strategy are iteratively adjusted. If the patient's BFDI begins to approach BFDI ref , then gradually extend the detection cycle; otherwise, further shorten the cycle and strengthen intervention. The formula for updating the detection cycle is as follows: ， In the formula, ΔSTI is the change of the latest trend indicator relative to the basic trend indicator, STI baseline is the reference trend indicator of the initial state, G is the sensitivity coefficient, T updated is the updated detection period.

8. A nursing system based on a hierarchical monitoring mode, used to implement a nursing method based on a hierarchical monitoring mode as described in any one of claims 1 to 7, characterized in that: It includes regular detection module, data collection module, feature extraction module, intelligent prediction module, state division module, routine monitoring module and dynamic adjustment module: The regular detection module, firstly, regularly detects the skin microcirculation perfusion status of severely burned patients according to the preset detection cycle to evaluate the tissue perfusion and wound recovery; The data acquisition module acquires the patient's skin microcirculation changes in real time during the test, and provides continuous and comprehensive raw data for subsequent analysis; The feature extraction module pre-processes the acquired data, and based on the pre-processed data, selects key features that reflect the trend of the patient's blood in the blood vessels toward hypercoagulation, and analyzes the extracted key features in detail within the detection window to quantify the trend of the patient's blood changes; The intelligent prediction module inputs the key features that have undergone quantitative analysis into a pre-trained deep learning model, and uses the deep learning model to make intelligent predictions on the patient's blood change trends; The state classification module divides the patient's blood change state into two categories: "blood hypercoagulability trend" and "blood rheological balance" based on the prediction results of the deep learning model; Routine monitoring module: For patients in a state of blood rheological balance, there is no need to increase the detection frequency, and routine care and monitoring can be carried out according to the preset cycle; The dynamic adjustment module dynamically increases the frequency of skin microcirculation change detection for patients in a state of blood hypercoagulation tendency, timely captures further microvascular obstruction characteristics, and prevents the patient's condition from worsening.

Images