Intelligent pulse anti-thrombus air pressure treatment system based on multi-point pressure monitoring

Abstract

The invention belongs to the technical field of air pressure treatment, and discloses an intelligent pulse anti-thrombus air pressure treatment system based on multi-point pressure monitoring. Comprising the following steps: constructing a pressure sensing network, and monitoring real-time pressure values of a gas tank, a left gas path and a right gas path; performing pressure comprehensive regionalization processing on the feet of the patient to obtain a patient foot regionalization model; constructing a pressure distribution model based on the patient foot regionalization model; according to the pressure distribution model, target pressure values of the left gas path and the right gas path are determined; in the first pressure output period, the gas tank is controlled to release pressure to the left gas path and the right gas path, and the pressure difference value is calculated; on the basis of the pressure difference value, the pressure compensation amount of the gas tank is adjusted through an improved iceberg algorithm, pressure compensation is conducted on the left gas path and the right gas path on the basis of the pressure compensation amount, and a difference value allowable range is preset till the pressure difference value is within the difference value allowable range; the discomfort caused by too high or too low pressure is effectively avoided, and the treatment experience of a patient is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pneumatic therapy, and more specifically, to an intelligent pulse anti-thrombotic pneumatic therapy system based on multi-point pressure monitoring. Background Art

[0002] Antithrombotic pneumatic compression therapy is a common clinical treatment method, which is widely used in the treatment of diseases such as varicose veins of the lower limbs and lymphedema of the lower limbs. This therapy promotes blood circulation and lymphatic return by applying a certain pressure to the patient's lower limbs, thereby achieving the purpose of reducing swelling and preventing varicose veins. The traditional antithrombotic pneumatic compression therapy system is usually composed of a gas tank, an air circuit and a pressurizing device. The gas tank provides compressed air for the entire system, and the air circuit transmits the compressed air to the pressurizing device, which is responsible for applying pressure to the patient's lower limbs. During the treatment process, medical staff need to set appropriate pressure parameters based on experience, and control the gas tank to release compressed air to the air circuit and the pressurizing device, thereby applying the required pressure to the patient's lower limbs.

[0003] However, the anti-thrombotic pneumatic pressure therapy system currently widely used in clinical practice has many shortcomings. Due to ignoring the individual differences of patients, these systems often take a one-size-fits-all approach when applying pressure, and fail to fully consider the specific shape characteristics of the patient's foot. This causes some patients to feel obvious discomfort during the treatment process, such as pain caused by excessive pressure, or local blood circulation problems caused by uneven pressure distribution. Especially for those patients with obvious foot deformations, it is even more difficult to achieve ideal treatment effects using conventional pressure schemes. On the other hand, the existing system also has many defects in pressure monitoring and regulation. The traditional pressure sensing network has limited coverage and poor connectivity, and cannot achieve comprehensive monitoring of the entire treatment area. Even if abnormal pressure is detected, there is a lack of effective compensation mechanism, and the output of the pneumatic system cannot be adjusted in time, resulting in a large deviation between the actual pressure applied to the patient and the ideal pressure, which seriously affects the quality of treatment.

[0004] In view of this, the present invention proposes an intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent pulse anti-thrombotic air pressure treatment system based on multi-point pressure monitoring, comprising: a sensor network module, a pressure comprehensive regionalization processing module, a distribution model building module, a control module, a gas tank, a left gas circuit and a right gas circuit; The sensor network module is used to construct a pressure sensor network, and use the pressure sensor network to monitor the real-time pressure values ​​of the gas tank, the left gas line and the right gas line; A pressure comprehensive regionalization processing module is used to perform pressure comprehensive regionalization processing on the patient's foot, divide the patient's foot into a number of grid units, and obtain a regionalized model of the patient's foot; A distribution model building module calculates the required pressure value for each grid unit based on the regionalized model of the patient's foot and builds a pressure distribution model; The control module is used to determine the target pressure values ​​of the left gas path and the right gas path according to the pressure distribution model; define two pressure output cycles; in the first pressure output cycle, control the gas tank to release the pressure to the left gas path and the right gas path, and monitor the real-time pressure value; Compare the real-time pressure value with the target pressure value and calculate the pressure difference. Based on the pressure difference, adjust the pressure compensation amount of the gas tank through the improved iceberg algorithm. Before the second pressure output cycle, perform pressure compensation on the left and right gas paths based on the pressure compensation amount, and preset the allowable range of difference until the pressure difference is within the allowable range of difference.

[0006] Furthermore, the pressure sensing network is configured as follows: Determine the area where pressure sensors need to be deployed, and record it as the area to be covered; initially dynamically deploy a fixed number of pressure sensor nodes in the area to be covered; draw a minimum circle around each pressure sensor node as the center, so that the minimum circle completely covers the sensing range of the corresponding pressure sensor node; divide each minimum circle into k1 equal fan-shaped areas; k1 is a pre-set integer constant; For each pressure sensor node, at least one non-empty sector area is selected from the k1 sector areas as the effective working area of ​​the corresponding pressure sensor node; a pressure sensor node is selected in each sector area as the key node of the corresponding sector area, responsible for pressure monitoring and data transmission in the sector area; the key nodes are interconnected through wireless communication links to form the backbone network of the pressure sensing network, and the backbone network is dynamically adjusted to obtain the final pressure sensing network.

[0007] Furthermore, the deployment method of the pressure sensor node includes: Collect historical pressure data of the area to be covered, perform two-dimensional discrete wavelet transform on the historical pressure data, and obtain a series of wavelet coefficients of different scales and directions; calculate the energy value of each wavelet coefficient, and use the energy value of the wavelet coefficient to plan the gradient distribution map of the pressure field in the area to be covered; Preset pressure threshold, based on which the gradient distribution of pressure field is Figure 2 The high gradient region and the low gradient region are obtained by valuing; the areas of the high gradient region and the low gradient region are obtained, and the expected density of the high gradient region is calculated based on the areas of the high gradient region and the low gradient region. and the desired density in low gradient regions ; ;in, is the total number of preset pressure sensor nodes, is the area of ​​high gradient region, is the total area of ​​the gradient distribution map, is a positive constant; is the average gradient value in the high gradient area, is the average residual energy of nodes in the high gradient region; is the preset node initial energy; ;in, is the area of ​​the low gradient region, is the average gradient value in the low gradient area, is the average residual energy of the nodes in the low gradient area; the suppressed Monte Carlo random process is used to deploy pressure sensor nodes in the high gradient area and the low gradient area respectively according to the calculated expected density.

[0008] Furthermore, the method of presetting the pressure threshold includes: The gradient distribution map is divided into several small areas b; each small area b is subjected to wavelet transform to obtain the wavelet coefficients of each scale of the small area ; is the scale number; Wavelet coefficients of various scales based on small regions , calculate the threshold value corresponding to the small area b ; ;in, To take the median, is the size of the small area b; is the length of the small area b, is the width of the small area b; Utilizing Thresholds For wavelet coefficients Perform threshold denoising and reconstruct the denoised small area b'; calculate the mean value of b' as the adaptive threshold; perform weighted average of the adaptive thresholds obtained from all small areas to obtain the pressure threshold.

[0009] Furthermore, the method of dynamically adjusting the path of the backbone network includes: The pressure sensor nodes that are not critical nodes are recorded as non-critical nodes; for each non-critical node v, a routing factor table is constructed, and the table items are (d', n', ξ) triples, where d' represents the critical node, n' represents the next hop node, and ξ represents the routing selection probability; initially, all ξ values ​​are equal; Each non-critical node v periodically releases artificial ants to the backbone network to simulate the process of forward exploration and backward propagation of artificial ants in the backbone network; forward exploration is: the artificial ant starts from v and randomly selects the next hop node according to probability ξ until it reaches a critical node d'; backward propagation is: the artificial ant carries the length of this path and the sequence of pressure sensor nodes and returns to v along the original path; then v discovers several paths to d'; for each discovered path , calculate its selection probability ; ;in, is a constant, is the length of the path, is the minimum remaining node energy on the path; and It is an adjustable parameter; The calculated Update the corresponding ξ value to the routing factor table; when the non-critical node needs to send pressure data to the critical node, a path is randomly selected according to the probability of ξ, and the pressure data is transmitted along this path; at the same time, the routing factor table is continuously updated.

[0010] Furthermore, the method of performing comprehensive regionalized pressure treatment on the patient's foot includes: Obtain the pressure distribution data at the contact point between the foot and the inner wall of the glove; define a three-dimensional discrete grid space, and randomly distribute a fixed number of live bacterial cells in the discrete grid space as the initial population; set the initial state for each live bacterial cell, the state includes life cycle, position coordinates and activity value; set the growth rules of the live bacterial cells, the growth rules include cell division rules, cell migration rules, cell aging rules, cell competition rules and apoptosis rules; The pressure distribution data is a two-dimensional matrix formed by the pressure value at the contact point between the foot and the inner wall measured by the pressure sensor, which is recorded as the pressure distribution data matrix; the pressure distribution data matrix is ​​interpolated in three dimensions to obtain the three-dimensional pressure field data F, the three-dimensional pressure field data F is mapped to the discrete grid space, the three-dimensional pressure field data F and the discrete grid space are normalized, and for each grid point, the pressure value of the corresponding position in the three-dimensional pressure field data F is searched; a high pressure threshold value Phigh and a low pressure threshold value Plow are set, and for each grid point, if the corresponding pressure value is greater than Phigh, the grid point is divided into a high pressure area; for each grid point, if the corresponding pressure value is less than Plow, the grid point is divided into a low pressure area; For the grid points classified as high pressure areas , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at The value of the life cycle randomly assigned at the beginning; To reduce the control parameters; is the grid point The pressure value at the corresponding position; For the grid points divided into low pressure areas , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at is the increment control parameter, is the grid point The pressure value at the corresponding position; Count the number Nhigh of all grid points divided into high-pressure areas, randomly select Nhigh grid points, and place live bacterial cells at these grid points according to the preset density as the initial seed points for growth; set an upper limit on the number of iterations, and at the beginning of each iteration, execute the growth rule for all surviving live bacterial cells. If the number of iterations reaches the upper limit, or the live bacterial cells have filled the entire foot area, terminate the iteration, and the final three-dimensional distribution of live bacterial cells is the preliminary patient foot regionalization model; The preliminary regionalized model of the patient's foot is smoothed and divided into regular grid units to obtain the regionalized model of the patient's foot.

[0011] Furthermore, the cell division rule is: When the current division age of the live bacterial cell reaches the preset division age threshold, and there are free grid points around the live bacterial cell for new cells to occupy, a new daughter cell is generated at the free grid point, and the initial state of the daughter cell inherits the live bacterial cell; and the age of the live bacterial cell is reset to 0; The rules for cell migration are: The current migration age of the live bacterial cell reaches a preset migration age threshold; all idle grid points around the live bacterial cell are obtained, and according to a preset migration probability, an idle grid point is randomly selected, and the position coordinates of the live bacterial cell are migrated to the selected grid point; The rules of cellular senescence are: The time step is preset, and the age of all surviving bacterial cells increases by 1 after each time step. When the age of a bacterial cell is greater than its life cycle, the bacterial cell enters the aging state, and the activity value of the aging bacteria gradually decreases until the activity value is 0. The cell competition rules are: Define a cell competition radius r. Competition will only occur between live bacterial cells within the cell competition radius. For each live bacterial cell c, count the number of other live bacterial cells g within its cell competition radius. Set a cell density threshold. When g is greater than the cell density threshold, the survival probability of live bacterial cell c at the current position is calculated based on the cell activity value. ; ;in, To control the parameters, is the current activity value of the living bacterial cells, is the preset activity threshold parameter; a random number in the range of [0, 1] is generated. If this random number is greater than , then the live bacterial cell c is marked as dying; The rules of apoptosis are: Define an apoptosis age threshold aop and an apoptosis probability pap. For each live bacterial cell c whose age is greater than aop, generate a random number in the range of [0, 1]. If this random number is less than pap, the live bacterial cell c is marked as dying. The live bacterial cells marked as dying will be completely removed in the next time step.

[0012] Furthermore, the method of calculating the pressure value required for each grid unit includes: The target area to apply pressure is selected and marked as a seed area. The seed area is scanned to obtain the grid cells on its surface. The scanning pressure value Pt is set. For the surface grid cells of the seed area, the pressure value is set to Pt. The direction perpendicular to the surface of the foot is defined as the scanning line direction. Starting from the grid cells on the surface of the seed area, the surface of the patient's foot regionalization model is moved forward along the scanning line direction until the next unassigned grid cell is encountered. The pressure value of the unassigned grid cell is set to an attenuation value of the current grid cell pressure value. Repeat until the scanning line encounters the boundary of the patient's foot regionalization model. Finally, the pressure value of each grid cell is obtained, that is, a pressure distribution model based on the patient's three-dimensional foot model is constructed.

[0013] Furthermore, in the pressure distribution model, the pressure values ​​of all grid cells in the left foot area and the right foot area are counted respectively. For the left foot area, the grid cell with the largest pressure value is found, recorded as Lmax, and the grid cell with the smallest pressure value is found, recorded as Lmin; the average pressure value Lavg of the left foot area is calculated, and the target pressure value of the left air path is set to Lmax or Lavg; for the right foot area R, the grid cell with the largest pressure value is found, recorded as Rmax, and the grid cell with the smallest pressure value is found, recorded as Rmin, the average pressure value Ravg of the right foot area is calculated, and the target pressure value of the right air path is set to Rmax or Ravg.

[0014] Furthermore, the method of adjusting the pressure compensation amount of the gas tank includes: Initialize the properties of the iceberg, including its volume ,thickness , area, drift rate, and melt rate, where area is volume divided by thickness; ;in, is the preset maximum volume value, is the differential pressure value, is the preset reference pressure difference, and are two volume mapping adjustment parameters, and are two correction parameters; ;in, is the preset minimum thickness value, is the preset maximum thickness value, Adjust parameters for thickness mapping; Initialize a group of icebergs, each of which corresponds to a pressure compensation amount, to obtain an iceberg group; iteratively optimize the iceberg group and update the drift rate and melting rate of each iceberg; the update formula of the drift rate is: ;in, is the drift speed of the iceberg, is the preset maximum drift rate, The current area of ​​the iceberg; is the preset maximum area; is the drift adjustment parameter; the update formula of the melting rate is: ;in, is the melting rate of the iceberg, is the preset maximum melting rate, is the melt adjustment parameter; Each iceberg drifts according to its own drift rate, and each iceberg melts according to its own melting rate; traverse the entire iceberg group, for each iceberg, compare its current volume V with the preset growth threshold Vth, if V is less than Vth, then mark the iceberg as an iceberg that needs to grow, if it is greater than or equal to Vth, then the iceberg does not need to grow temporarily; For icebergs marked as needing to grow, define their growth amount ;in, is a growth factor; Add it to its current volume V to get the new volume; After each iteration, the cold value of each iceberg is calculated; the cold value ; until the maximum number of iterations is reached, the iceberg with the largest cold value is selected as the current optimal solution, that is, the current optimal pressure compensation amount is obtained.

[0015] The technical effects and advantages of the intelligent pulse anti-thrombotic pneumatic pressure treatment system based on multi-point pressure monitoring of the present invention are as follows: the present invention can provide patients with personalized and accurate anti-thrombotic pneumatic pressure treatment plans, greatly improving the effectiveness and comfort of treatment; by performing comprehensive regional pressure treatment on the patient's foot, the shape characteristics of the foot are fully captured, and combined with a reasonable pressure distribution model, a pressure application plan that meets the individual needs of the patient is tailored; this personalized treatment method avoids the drawbacks of the traditional one-size-fits-all approach, ensuring that every patient can obtain the best pressure regulation effect, thereby maximizing blood circulation and accelerating recovery from the disease; at the same time, by adopting pressure monitoring and compensation technology, by building a high-coverage, high-connectivity pressure sensing network, the pressure changes during the treatment process are monitored in real time, and based on the improved optimization algorithm, the pressure output of the pneumatic system is dynamically adjusted to ensure that the pressure applied to the patient is always maintained within the ideal target range, effectively avoiding the discomfort caused by excessively high or low pressure, and greatly improving the patient's treatment experience; in addition, the pressure distribution model and compensation mechanism can timely adjust the pressure parameters according to the patient's specific condition and feedback, realize refined and dynamic pressure control, and further optimize the treatment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring of the present invention; Figure 2 It is a schematic diagram of the intelligent pulse anti-thrombotic air pressure treatment method based on multi-point pressure monitoring of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1 See also Figure 1 As shown, the intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring described in this embodiment includes: a sensor network module, a pressure comprehensive regionalization processing module, a distribution model building module, a control module, a gas tank, a left gas circuit and a right gas circuit; The sensor network module is used to construct a pressure sensor network, and use the pressure sensor network to monitor the real-time pressure values ​​of the gas tank, the left gas line and the right gas line; A pressure comprehensive regionalization processing module is used to perform pressure comprehensive regionalization processing on the patient's foot, divide the patient's foot into a number of grid units, and obtain a regionalized model of the patient's foot; A distribution model building module calculates the required pressure value for each grid unit based on the regionalized model of the patient's foot and builds a pressure distribution model; The control module is used to determine the target pressure values ​​of the left gas path and the right gas path according to the pressure distribution model; define two pressure output cycles; in the first pressure output cycle, control the gas tank to release the pressure to the left gas path and the right gas path, and monitor the real-time pressure value; Compare the real-time pressure value with the target pressure value and calculate the pressure difference. Based on the pressure difference, adjust the pressure compensation amount of the gas tank through the improved iceberg algorithm. Before the second pressure output cycle, perform pressure compensation on the left and right gas paths based on the pressure compensation amount, and preset the allowable range of difference until the pressure difference is within the allowable range of difference.

[0019] The pressure sensing network is configured as follows: Determine the area where pressure sensors need to be arranged, and record it as the area to be covered; such as gas tanks, left-path gas paths and right-path gas paths, etc.; initially dynamically deploy a fixed number of pressure sensor nodes in the area to be covered; specifically, collect historical pressure data of the area to be covered, perform two-dimensional discrete wavelet transform on the historical pressure data, and obtain a series of wavelet coefficients of different scales and directions; calculate the energy value of each wavelet coefficient, which reflects the change characteristics of the pressure data in the scale direction; use the energy value of the wavelet coefficient to plan the gradient distribution map of the pressure field in the area to be covered.

[0020] Preset the pressure threshold. Specifically, divide the gradient distribution map into several small areas b; perform wavelet transform on each small area b to obtain the wavelet coefficients of each scale of the small area. ; is the scale number; Wavelet coefficients of various scales based on small regions , calculate the threshold value corresponding to the small area b ; ;in, To take the median, is the size of the small area b; is the length of the small area b, is the width of the small area b; Utilizing Thresholds For wavelet coefficients Perform threshold denoising and reconstruct the denoised small area b'; calculate the mean value of b' as the adaptive threshold; perform weighted average of the adaptive thresholds obtained from all small areas to obtain the pressure threshold.

[0021] The gradient distribution of the pressure field is based on the pressure threshold Figure 2 The high gradient area and the low gradient area are obtained by quantization; the high gradient area corresponds to the key area where the pressure changes dramatically; the area of ​​the high gradient area and the low gradient area is obtained, and the expected density of the high gradient area is calculated based on the area of ​​the high gradient area and the low gradient area. and the desired density in low gradient regions ; ;in, is the total number of preset pressure sensor nodes, is the area of ​​high gradient region, is the total area of ​​the gradient distribution map, is a positive constant used to control the influence of the gradient value on the gain; is the average gradient value in the high gradient area, reflecting the severity of the pressure change in the area. is the average residual energy of nodes in the high gradient area. The higher the residual energy, the stronger the survivability of the nodes in the area, and the node density can be appropriately increased; is the preset node initial energy, which is a positive constant; ;in, is the area of ​​the low gradient region, is the average gradient value in the low gradient area, is the average residual energy of the nodes in the low gradient area; the suppressed Monte Carlo random process is used to deploy pressure sensor nodes in the high gradient area and the low gradient area according to the calculated expected density to achieve non-uniform distribution.

[0022] With each pressure sensor node as the center, draw a minimum circle around it so that the minimum circle completely covers the sensing range of the corresponding pressure sensor node; divide each minimum circle into k1 equal sector areas; k1 is a pre-set integer constant, usually 6; for each pressure sensor node, select at least one non-empty sector area from the k1 sector areas as the effective working area of ​​the corresponding pressure sensor node.

[0023] In each sector area, a pressure sensor node is selected as the key node of the corresponding sector area, which is responsible for pressure monitoring and data transmission in the sector area; the key nodes are interconnected through wireless communication links to form the backbone network of the pressure sensing network, and the backbone network is dynamically adjusted to obtain the final pressure sensing network.

[0024] The pressure sensor nodes that are not critical nodes are recorded as non-critical nodes; for each non-critical node v, a routing factor table is constructed, and the table items are (d', n', ξ) triples, where d' represents the critical node, n' represents the next hop node, and ξ represents the routing selection probability; initially, all ξ values ​​are equal.

[0025] Each non-critical node v periodically releases artificial ants to the backbone network to simulate the process of forward exploration and backward propagation of artificial ants in the backbone network; Forward exploration is as follows: the artificial ant starts from v and randomly selects the next hop node according to probability ξ until it reaches a key node d'; backward propagation is as follows: the artificial ant carries the length of this path and the sequence of pressure sensor nodes and returns to v along the original path; then v finds several paths leading to d'.

[0026] For each discovered path , calculate its selection probability ; ;in, is a constant representing the path quality weight, is the length of the path (number of path nodes), is the minimum remaining node energy on the path (for a path from a non-critical node to a critical node, the remaining energy value of the node with the smallest remaining energy among all the nodes on the path is estimated based on the working status and power consumption model of the node); and is an adjustable parameter that controls the weight of length and energy.

[0027] The calculated Update to the corresponding ξ value of the routing factor table; when the non-critical node needs to send pressure data to the critical node, a path is randomly selected according to the probability of ξ, and the pressure data is transmitted along this path; at the same time, the routing factor table is continuously updated; after the critical node receives repeated pressure data from different paths, it performs data fusion to obtain the final pressure data; Through this hierarchical network construction method, the entire monitoring area can be guaranteed to be fully covered, and a connected sensor network topology with high coverage can be quickly constructed.

[0028] The methods of comprehensive regional pressure treatment of the patient's foot include: A detachable foot cover is set up, and the interior of the foot cover is composed of several pressure sensor arrays. The patient is asked to put both feet into the foot cover so that the feet are completely in contact with the inner wall of the foot cover; the pressure sensor array in the foot cover is activated to obtain the pressure distribution data of the contact point between the foot and the inner wall of the foot cover.

[0029] Define a three-dimensional discrete grid space, the size of which is set according to the foot area to be reconstructed; randomly distribute a fixed number of live bacterial cells in the discrete grid space as the initial population; set the initial state for each live bacterial cell, which includes life cycle, position coordinates and activity value; randomly assign a life cycle value to each cell, ranging between the preset minimum and maximum life cycles, such as [10, 50]; randomly select an idle grid point in the discrete grid space as the initial position coordinate of the live bacterial cell; initially, all live bacterial cells are in an active state, and the activity value is set to 1.

[0030] Set the growth rules of living bacterial cells, including cell division rules, cell migration rules, cell aging rules, cell competition rules and apoptosis rules; The rules for cell division are: The current division age of the living bacterial cells reaches the preset division age threshold, where division age = total life cycle × division coefficient, such as 0.6; If there are idle grid points around the living bacterial cell (26 neighborhoods) for the new cell to occupy, a new daughter cell is generated at the idle grid point, and the initial state of the daughter cell inherits the living bacterial cell; and the age of the living bacterial cell is reset to 0.

[0031] The rules for cell migration are: The current migration age of the live bacterial cell reaches the preset migration age threshold, where migration age = total life cycle × migration coefficient, such as 0.3; all idle grid points around the live bacterial cell (26 neighborhoods) are obtained, and an idle grid point is randomly selected according to the preset migration probability, and the position coordinates of the live bacterial cell are migrated to the selected grid point.

[0032] The rules of cellular senescence are: The time step is preset, and the age of all surviving bacterial cells increases by 1 after each time step; when the age of a live bacterial cell is greater than its life cycle, the live bacterial cell enters the aging state, and the activity value of the live bacteria in the aging state gradually decreases (multiplied by a value of 0-1) until it is completely inactivated, that is, the activity value is 0.

[0033] The cell competition rules are: Define a cell competition radius r. Competition will only occur between live bacterial cells within the range of this cell competition radius. For each live bacterial cell c, count the number of other live bacterial cells g within its cell competition radius. Set a cell density threshold. When g is greater than the cell density threshold, it is considered that the area where the live bacterial cell c is located is too crowded. According to the cell activity value, calculate the survival probability of the live bacterial cell c at the current location. ; ;in, is a control parameter used to control the degree of mapping. is the current activity value of the living bacterial cells, is a preset activity threshold parameter; Generate a random number in the range [0, 1]. If the random number is greater than , then the live bacterial cell c is marked as dying.

[0034] The rules of apoptosis are: Define an apoptosis age threshold aop and an apoptosis probability pap. For each live bacterial cell c whose age is greater than aop, generate a random number in the range of [0, 1]. If this random number is less than pap, the live bacterial cell c is marked as dying. The live bacterial cells marked as dying will be completely removed in the next time step.

[0035] By setting the initial state and growth rules, a dynamic model that can simulate the behavior of biological cells can be constructed, laying the foundation for reconstructing the regionalized model of the patient's foot.

[0036] The pressure distribution data is a two-dimensional matrix composed of the pressure values ​​at the contact point between the foot and the inner wall measured by the pressure sensor, which is recorded as the pressure distribution data matrix. The pressure distribution data matrix is ​​three-dimensionally interpolated to obtain the three-dimensional pressure field data F. The interpolation method can be nearest neighbor interpolation, trilinear interpolation or Gaussian kernel interpolation.

[0037] The three-dimensional pressure field data F is mapped to the discrete grid space, and the three-dimensional pressure field data F and the discrete grid space are normalized to make the two have the same size, and for each grid point, the pressure value of the corresponding position in the three-dimensional pressure field data F is searched; a high pressure threshold value Phigh and a low pressure threshold value Plow are set, and for each grid point, if the corresponding pressure value is greater than Phigh, the grid point is divided into a high pressure area; for each grid point, if the corresponding pressure value is less than Plow, the grid point is divided into a low pressure area.

[0038] For the grid points classified as high pressure areas , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at The value of the life cycle randomly assigned at the beginning; To reduce the control parameters, it is used to control the life cycle reduction rate and is adjusted according to the actual situation; is the grid point The pressure value at the corresponding position.

[0039] For the grid points divided into low pressure areas , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at It is an increment control parameter, which is used to control the rate of increase of the life cycle. is the grid point The pressure value at the corresponding position; the number Nhigh of all grid points divided into high pressure areas is counted, Nhigh grid points are randomly selected, and living bacterial cells are placed at these grid points according to a preset density as initial seed points for growth.

[0040] By embedding pressure data into the model, high-pressure areas will serve as seeds and restriction areas for the growth of living bacteria, ultimately generating a model that can reflect the three-dimensional shape of the foot.

[0041] An upper limit on the number of iterations is set. At the beginning of each iteration, the growth rule is executed for all surviving bacterial cells. If the number of iterations reaches the upper limit or the live bacterial cells have filled the entire foot area, the iteration is terminated. The final three-dimensional distribution of live bacterial cells is the preliminary patient foot regionalization model. The preliminary patient foot regionalization model is smoothed (such as Laplace smoothing, Gaussian smoothing, etc.) to eliminate noise and discontinuity, and is discretized and divided into regular grid units. Each grid unit corresponds to a small area of ​​the foot, and the patient foot regionalization model is obtained.

[0042] The required pressure value for each grid cell is calculated by: Select the target area to apply pressure, such as the ankle, and mark it as the seed area. Scan the seed area to obtain the grid cells on its surface. Set the scanning pressure value Pt. For the surface grid cells of the seed area, set its pressure value to Pt. Define the direction perpendicular to the surface of the foot as the scanning line direction. Starting from the grid cells on the surface of the seed area, move along the surface of the patient's foot regionalization model along the scanning line direction until encountering the next unassigned grid cell. Set the pressure value of the unassigned grid cell to an attenuation value of the current grid cell pressure value, such as 90% of the current value. Repeat until the scanning line encounters the boundary of the patient's foot regionalization model. For some special areas, such as the sole and arch, a higher target pressure value can be set in advance to give these areas greater pressure. Finally, the pressure value of each grid cell is obtained, that is, a pressure distribution model based on the patient's three-dimensional foot model is constructed.

[0043] In pulse antithrombotic pneumatic pressure therapy, the pressure distribution model is used to guide the pneumatic system to apply appropriate pressure to the patient's feet. Specifically, the left and right air circuits respectively control the pressure applied to the patient's foot area. According to the pressure distribution model, the target pressure values ​​of the left and right air circuits can be determined, thereby achieving precise pressure regulation of the patient's feet. During the treatment, the pressure values ​​of the left and right air circuits are monitored in real time through the pressure sensing network and compared with the target pressure values ​​for pressure compensation, thereby ensuring that the pressure applied to the patient meets expectations. According to the individual foot shape characteristics of the patient, a reasonable pressure distribution plan is tailored for them to improve the pertinence and effectiveness of antithrombotic pneumatic pressure therapy.

[0044] In the pressure distribution model, the pressure values ​​of all grid cells in the left foot area and the right foot area are counted respectively. For the left foot area, the grid cell with the largest pressure value is found, recorded as Lmax, and the grid cell with the smallest pressure value is found, recorded as Lmin. The average pressure value Lavg of the left foot area is calculated, and the target pressure value of the left airway is set to Lmax or Lavg, depending on the treatment needs. If Lmax is used as the target pressure value, it can ensure that the corresponding area obtains sufficient pressure. If Lavg is used as the target pressure value, the entire left foot area can obtain an average reasonable pressure.

[0045] For the right foot area R, find the grid unit with the largest pressure value, recorded as Rmax, find the grid unit with the smallest pressure value, recorded as Rmin, calculate the average pressure value Ravg of the right foot area, and the target pressure value of the right air path can be set to Rmax or Ravg, and the determination method is the same as the left path; if there are abnormal points with too high or too low pressure values, their influence can be eliminated by removing the outliers; after the target pressure values ​​of the left and right air paths are determined, they can be compared with the real-time pressure values ​​respectively to perform pressure compensation, thereby achieving precise pressure control.

[0046] Ideally, the pressure compensation amount is equal to the opposite of the pressure difference, so that the real-time pressure value after compensation is exactly equal to the target pressure value. However, in actual operation, due to the influence of various uncertain factors, the pressure compensation amount usually cannot completely offset the pressure difference and can only minimize the pressure difference.

[0047] Ways to adjust the pressure compensation of the gas tank include: Initialize the properties of the iceberg, including its volume ,thickness , area (the area of ​​the iceberg above the water, obtained by dividing its volume by its thickness), drift rate, and melt rate (the rate at which the iceberg melts); the larger the volume, the higher the probability of this pressure compensation.

[0048] ;in, is the preset maximum volume value, is the differential pressure value, is the preset reference pressure difference, and There are two volume mapping adjustment parameters, which are used to control the mapping relationship between volume and pressure difference. and There are two correction parameters used to correct the volume according to the size of the pressure difference; ;in, is the preset minimum thickness value, is the preset maximum thickness value, Adjusting parameters for thickness mapping to control the mapping relationship between thickness and pressure difference; Initialize a group of icebergs, each of which corresponds to a pressure compensation amount, to obtain an iceberg group; iteratively optimize the iceberg group and update the drift rate and melting rate of each iceberg; the update formula of the drift rate is: ;in, is the drift speed of the iceberg, is the preset maximum drift rate, The current area of ​​the iceberg; is the preset maximum area; It is a drift adjustment parameter used to control the mapping relationship between drift rate and region size.

[0049] The updated formula for the melting rate is: ;in, is the melting rate of the iceberg, is the preset maximum melting rate, is a melting adjustment parameter that controls the mapping between melting rate and region size; Each iceberg drifts according to its own drift rate, simulating the effect of ocean currents. Each iceberg melts according to its own melting rate, and its volume gradually decreases.

[0050] Traverse the entire iceberg group, and for each iceberg, compare its current volume V with the preset growth threshold Vth. If V is less than Vth, the iceberg is marked as an iceberg that needs to grow. If it is greater than or equal to Vth, the iceberg does not need to grow temporarily. For the iceberg marked as needing to grow, define its growth amount ;in, is the growth factor, a number in the interval (0, 1); the growth amount Add it to its current volume V to get a new volume, simulating the growth of an iceberg in a cold environment.

[0051] After each iteration, the cold value of each iceberg is calculated to evaluate its quality as a pressure compensation quantity; the cold value ; until the maximum number of iterations is reached, the iceberg with the largest cold value is selected as the current optimal solution, that is, the current optimal pressure compensation amount is obtained.

[0052] After obtaining the final optimal pressure compensation amount, before the start of the second pressure output cycle, the control gas tank performs corresponding pressure compensation on the left and right gas circuits, so that the real-time pressure value gradually approaches the target pressure value.

[0053] By simulating the drifting, melting and growth process of icebergs in nature, we can effectively search for the optimal solution in the global and local range, so as to obtain a reasonable amount of pressure compensation and ensure that the pressure applied to the patient during treatment is accurately controllable.

[0054] This embodiment can provide patients with personalized and accurate anti-thrombotic air pressure treatment plans, greatly improving the effectiveness and comfort of treatment; by performing comprehensive regional pressure treatment on the patient's foot, the shape characteristics of the foot are fully captured, and combined with a reasonable pressure distribution model, a pressure application plan that meets the individual needs of the patient is tailored; this personalized treatment method avoids the drawbacks of the traditional one-size-fits-all approach, ensuring that every patient can obtain the best pressure regulation effect, thereby maximizing blood circulation and accelerating recovery; at the same time, by using pressure monitoring and compensation technology, by building a high-coverage, high-connectivity pressure sensing network, the pressure changes during the treatment process are monitored in real time, and based on the improved optimization algorithm, the pressure output of the air pressure system is dynamically adjusted to ensure that the pressure applied to the patient is always maintained within the ideal target range, effectively avoiding the discomfort caused by excessive or low pressure, and greatly improving the patient's treatment experience; in addition, the pressure distribution model and compensation mechanism can adjust the pressure parameters in time according to the patient's specific condition and feedback, realize refined and dynamic pressure control, and further optimize the treatment effect.

[0055] Example 2 See also Figure 2 As shown, the part not described in detail in this embodiment can be found in the description of embodiment 1, which provides an intelligent pulse anti-thrombotic air pressure treatment method based on multi-point pressure monitoring, including: Step 1: construct a pressure sensing network, and use the pressure sensing network to monitor the real-time pressure values ​​of the gas tank, the left gas path, and the right gas path; Step 2: Perform comprehensive regionalization of pressure on the patient's foot, divide the patient's foot into a number of grid units, and obtain a regionalized model of the patient's foot; Step 3: Based on the regionalized model of the patient's foot, the required pressure value of each grid unit is calculated, and a pressure distribution model is constructed; Step 4: Determine the target pressure values ​​of the left gas path and the right gas path according to the pressure distribution model; define two pressure output cycles; in the first pressure output cycle, control the gas tank to release pressure to the left gas path and the right gas path, and monitor the real-time pressure value; Compare the real-time pressure value with the target pressure value and calculate the pressure difference. Based on the pressure difference, adjust the pressure compensation amount of the gas tank through the improved iceberg algorithm. Before the second pressure output cycle, perform pressure compensation on the left and right gas paths based on the pressure compensation amount, and preset the allowable range of difference until the pressure difference is within the allowable range of difference.

[0056] Example 3 This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the intelligent pulse anti-thrombotic pneumatic pressure therapy method based on multi-point pressure monitoring provided above is implemented.

[0057] Since the electronic device introduced in this embodiment is an electronic device used to implement the intelligent pulse anti-thrombotic air pressure treatment method based on multi-point pressure monitoring in the embodiment of the present application, based on the intelligent pulse anti-thrombotic air pressure treatment method based on multi-point pressure monitoring introduced in the embodiment of the present application, the technical personnel of the field can understand the specific implementation of the electronic device of the present embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is not described in detail here. As long as the technical personnel of the field implement the electronic device used in the intelligent pulse anti-thrombotic air pressure treatment method based on multi-point pressure monitoring in the embodiment of the present application, it belongs to the scope of protection of the present application.

[0058] 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 and thresholds in the formula are set by technicians in this field according to actual conditions.

[0059] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. Intelligent pulse anti-thrombotic air pressure therapy system based on multi-point pressure monitoring, characterized in that: include: A sensor network module, a pressure integrated regional processing module, a distribution model building module, a control module, a gas tank, a left gas path and a right gas path; The sensor network module is used to construct a pressure sensor network, and use the pressure sensor network to monitor the real-time pressure values ​​of the gas tank, the left gas line and the right gas line; A pressure comprehensive regionalization processing module is used to perform pressure comprehensive regionalization processing on the patient's foot, divide the patient's foot into a number of grid units, and obtain a regionalized model of the patient's foot; A distribution model building module calculates the required pressure value for each grid unit based on the regionalized model of the patient's foot and builds a pressure distribution model; The control module is used to determine the target pressure values ​​of the left gas path and the right gas path according to the pressure distribution model; define two pressure output cycles; in the first pressure output cycle, control the gas tank to release the pressure to the left gas path and the right gas path, and monitor the real-time pressure value; Compare the real-time pressure value with the target pressure value and calculate the pressure difference; Based on the pressure difference, the pressure compensation amount of the gas tank is adjusted by improving the iceberg algorithm. Before the second pressure output cycle, the left and right gas paths are pressure compensated based on the pressure compensation amount, and the allowable range of difference is preset until the pressure difference is within the allowable range of difference.

2. The intelligent pulse antithrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 1 is characterized in that: The pressure sensing network is configured as follows: Determine the area where pressure sensors need to be deployed, and record it as the area to be covered; initially dynamically deploy a fixed number of pressure sensor nodes in the area to be covered; draw a minimum circle around each pressure sensor node as the center, so that the minimum circle completely covers the sensing range of the corresponding pressure sensor node; divide each minimum circle into k1 equal fan-shaped areas; k1 is a pre-set integer constant; For each pressure sensor node, at least one non-empty sector area is selected from the k1 sector areas as the effective working area of ​​the corresponding pressure sensor node; a pressure sensor node is selected in each sector area as the key node of the corresponding sector area, responsible for pressure monitoring and data transmission in the sector area; the key nodes are interconnected through wireless communication links to form the backbone network of the pressure sensing network, and the backbone network is dynamically adjusted to obtain the final pressure sensing network.

3. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 2 is characterized in that: The deployment method of the pressure sensor node includes: Collect historical pressure data of the area to be covered, perform two-dimensional discrete wavelet transform on the historical pressure data, and obtain a series of wavelet coefficients of different scales and directions; calculate the energy value of each wavelet coefficient, and use the energy value of the wavelet coefficient to plan the gradient distribution map of the pressure field in the area to be covered; A pressure threshold is preset, and the gradient distribution map of the pressure field is binarized based on the pressure threshold to obtain high gradient areas and low gradient areas; the areas of the high gradient areas and low gradient areas are obtained, and the expected density of the high gradient areas is calculated based on the areas of the high gradient areas and low gradient areas. and the desired density in low gradient regions ; ;in, is the total number of preset pressure sensor nodes, is the area of ​​high gradient region, is the total area of ​​the gradient distribution map, is a positive constant; is the average gradient value in the high gradient area, is the average residual energy of nodes in the high gradient region; is the preset node initial energy; ;in, is the area of ​​the low gradient region, is the average gradient value in the low gradient area, is the average residual energy of the nodes in the low gradient area; the suppressed Monte Carlo random process is used to deploy pressure sensor nodes in the high gradient area and the low gradient area respectively according to the calculated expected density.

4. The intelligent pulse antithrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 3 is characterized in that: The method of presetting the pressure threshold includes: The gradient distribution map is divided into several small areas b; each small area b is subjected to wavelet transform to obtain the wavelet coefficients of each scale of the small area ; is the scale number; Wavelet coefficients of various scales based on small regions , calculate the threshold value corresponding to the small area b ; ;in, To take the median, is the size of the small area b; is the length of the small area b, is the width of the small area b; Utilizing Thresholds For wavelet coefficients Perform threshold denoising and reconstruct the denoised small area b'; calculate the mean value of b' as the adaptive threshold; perform weighted average of the adaptive thresholds obtained from all small areas to obtain the pressure threshold.

5. The intelligent pulse antithrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 4 is characterized in that: The method of dynamically adjusting the path of the backbone network includes: The pressure sensor nodes that are not critical nodes are recorded as non-critical nodes; for each non-critical node v, a routing factor table is constructed, and the table items are (d', n', ξ) triples, where d' represents the critical node, n' represents the next hop node, and ξ represents the routing selection probability; initially, all ξ values ​​are equal; Each non-critical node v periodically releases artificial ants to the backbone network to simulate the process of forward exploration and backward propagation of artificial ants in the backbone network; forward exploration is: the artificial ant starts from v and randomly selects the next hop node according to probability ξ until it reaches a critical node d'; backward propagation is: the artificial ant carries the length of this path and the sequence of pressure sensor nodes and returns to v along the original path; then v discovers several paths to d'; for each discovered path , calculate its selection probability ; ;in, is a constant, is the length of the path, is the minimum remaining node energy on the path; and It is an adjustable parameter; The calculated Update the corresponding ξ value to the routing factor table; when the non-critical node needs to send pressure data to the critical node, a path is randomly selected according to the probability of ξ, and the pressure data is transmitted along this path; at the same time, the routing factor table is continuously updated.

6. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 5 is characterized in that: The method of performing comprehensive regional pressure treatment on the patient's foot includes: Obtain the pressure distribution data at the contact point between the foot and the inner wall of the glove; define a three-dimensional discrete grid space, and randomly distribute a fixed number of live bacterial cells in the discrete grid space as the initial population; set the initial state for each live bacterial cell, the state includes life cycle, position coordinates and activity value; set the growth rules of the live bacterial cells, the growth rules include cell division rules, cell migration rules, cell aging rules, cell competition rules and apoptosis rules; The pressure distribution data is a two-dimensional matrix composed of the pressure values ​​at the contact point between the foot and the inner wall measured by the pressure sensor, which is recorded as the pressure distribution data matrix; the pressure distribution data matrix is ​​interpolated in three dimensions to obtain the three-dimensional pressure field data F, the three-dimensional pressure field data F is mapped to the discrete grid space, the three-dimensional pressure field data F and the discrete grid space are normalized, and for each grid point, the pressure value of the corresponding position in the three-dimensional pressure field data F is searched; a high pressure threshold value Phigh and a low pressure threshold value Plow are set, for each grid point, if the corresponding pressure value is greater than Phigh, the grid point is divided into a high pressure area; for each grid point, if the corresponding pressure value is less than Plow, the grid point is divided into a low pressure area; for the grid points divided into the high pressure area , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at The value of the life cycle randomly assigned at the beginning; To reduce the control parameters; is the grid point The pressure value at the corresponding position; For the grid points divided into low pressure areas , the life cycle of the living bacterial cells at the grid point is adjusted to ;in, is the grid point The adjusted life span of viable cells at is the increment control parameter, is the grid point The pressure value at the corresponding position; Count the number Nhigh of all grid points divided into high-pressure areas, randomly select Nhigh grid points, and place live bacterial cells at these grid points according to the preset density as the initial seed points for growth; set an upper limit on the number of iterations, and at the beginning of each iteration, execute the growth rule for all surviving live bacterial cells. If the number of iterations reaches the upper limit, or the live bacterial cells have filled the entire foot area, terminate the iteration, and the final three-dimensional distribution of live bacterial cells is the preliminary patient foot regionalization model; The preliminary regionalized model of the patient's foot is smoothed and divided into regular grid units to obtain the regionalized model of the patient's foot.

7. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 6, characterized in that: The cell division rule is: When the current division age of the live bacterial cell reaches the preset division age threshold, and there are free grid points around the live bacterial cell for new cells to occupy, a new daughter cell is generated at the free grid point, and the initial state of the daughter cell inherits the live bacterial cell; And reset the age of living bacterial cells to 0; The cell migration rule is: the current migration age of the live bacterial cell reaches the preset migration age threshold; Obtain all the free grid points around the live bacterial cells, randomly select a free grid point according to the preset migration probability, and migrate the position coordinates of the live bacterial cells to the selected grid point; The cell aging rule is: preset time step, after each time step, the age of all surviving bacterial cells increases by 1; when the age of a bacterial cell is greater than its life cycle, the bacterial cell enters the aging state, and the activity value of the aging state gradually decreases until the activity value is 0; The cell competition rule is as follows: define a cell competition radius r, and competition will only occur between live bacterial cells within the range of the cell competition radius. For each live bacterial cell c, count the number of other live bacterial cells g within its cell competition radius; set a cell density threshold, and when g is greater than the cell density threshold, calculate the survival probability of the live bacterial cell c at the current position based on the cell activity value. ; ;in, To control the parameters, is the current activity value of the living bacterial cells, is a preset activity threshold parameter; Generate a random number in the range [0, 1]. If the random number is greater than , then the live bacterial cell c is marked as dying; The apoptosis rule is: define an apoptosis age threshold aop and an apoptosis probability pap. For each live bacterial cell c whose age is greater than aop, generate a random number in the range of [0, 1]. If this random number is less than pap, the live bacterial cell c is marked as dying. The live bacterial cells marked as dying will be completely removed in the next time step.

8. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 7, characterized in that: The method of calculating the pressure value required for each grid unit includes: The target area to apply pressure is selected and marked as a seed area. The seed area is scanned to obtain the grid cells on its surface. The scanning pressure value Pt is set. For the surface grid cells of the seed area, the pressure value is set to Pt. The direction perpendicular to the surface of the foot is defined as the scanning line direction. Starting from the grid cells on the surface of the seed area, the surface of the patient's foot regionalization model is moved forward along the scanning line direction until the next unassigned grid cell is encountered. The pressure value of the unassigned grid cell is set to an attenuation value of the current grid cell pressure value. Repeat until the scanning line encounters the boundary of the patient's foot regionalization model. Finally, the pressure value of each grid cell is obtained, that is, a pressure distribution model based on the patient's three-dimensional foot model is constructed.

9. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 8, characterized in that: In the pressure distribution model, the pressure values ​​of all grid cells in the left foot area and the right foot area are counted respectively. For the left foot area, the grid cell with the largest pressure value is found, recorded as Lmax, and the grid cell with the smallest pressure value is found, recorded as Lmin; the average pressure value Lavg of the left foot area is calculated, and the target pressure value of the left gas path is set to Lmax or Lavg; for the right foot area R, the grid cell with the largest pressure value is found, recorded as Rmax, and the grid cell with the smallest pressure value is found, recorded as Rmin, the average pressure value Ravg of the right foot area is calculated, and the target pressure value of the right gas path is set to Rmax or Ravg.

10. The intelligent pulse anti-thrombotic pneumatic pressure therapy system based on multi-point pressure monitoring according to claim 9, characterized in that: The method of adjusting the pressure compensation amount of the gas tank includes: Initialize the properties of the iceberg, including its volume ,thickness , area, drift rate, and melt rate, where area is volume divided by thickness; ;in, is the preset maximum volume value, is the differential pressure value, is the preset reference pressure difference, and are two volume mapping adjustment parameters, and are two correction parameters; ;in, is the preset minimum thickness value, is the preset maximum thickness value, Adjust parameters for thickness mapping; Initialize a group of icebergs, each of which corresponds to a pressure compensation amount, to obtain an iceberg group; iteratively optimize the iceberg group and update the drift rate and melting rate of each iceberg; the update formula of the drift rate is: ;in, is the drift speed of the iceberg, is the preset maximum drift rate, The current area of ​​the iceberg; is the preset maximum area; is the drift adjustment parameter; the update formula of the melting rate is: ;in, is the melting rate of the iceberg, is the preset maximum melting rate, is the melt adjustment parameter; Each iceberg drifts according to its own drift rate, and each iceberg melts according to its own melting rate; traverse the entire iceberg group, for each iceberg, compare its current volume V with the preset growth threshold Vth, if V is less than Vth, then mark the iceberg as an iceberg that needs to grow, if it is greater than or equal to Vth, then the iceberg does not need to grow temporarily; For icebergs marked as needing to grow, define their growth amount ;in, is a growth factor; Add it to its current volume V to get the new volume; After each iteration, the cold value of each iceberg is calculated; the cold value ; until the maximum number of iterations is reached, the iceberg with the largest cold value is selected as the current optimal solution, that is, the current optimal pressure compensation amount is obtained.

Images