Three-motor diagnosis and treatment bed angle adjusting method and system based on intelligent somatosensory

Abstract

The invention relates to the technical field of intelligent control, and discloses a three-motor diagnosis and treatment bed angle adjusting method and system based on intelligent somatosensory, and the method comprises the steps: obtaining pressure distribution data of each part of a patient body on a bed surface, posture data of the patient body and diagnosis and treatment demand data; based on the pressure distribution data and the attitude data, constructing a first equation to obtain a first somatosensory feature and obtaining a second somatosensory feature through a second equation; fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature; matching the first somatosensory feature, the second somatosensory feature and the third somatosensory feature with the diagnosis and treatment demand data through a preset intelligent mapping model, and converting into a mapping relationship between the somatosensory features and diagnosis and treatment demands to obtain mapping relationship data; and converting the mapping relation data to obtain an angle adjustment instruction, and controlling a driving motor to perform angle adjustment on the bed according to the angle adjustment instruction. According to the method, secondary injury caused by movement of the patient in the operation process is avoided.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an angle adjustment method and system for a three-motor diagnostic bed based on intelligent body sensing. Background Art

[0002] At present, with the continuous advancement of medical technology, the precise adjustment of the patient's position during surgery has become increasingly important. During surgery, patients are usually in an anesthetized state and completely lose the ability to move independently. Due to their weak physical condition, any improper movement may cause secondary harm to the patient. At the same time, the selection and adjustment of the surgical position is not only related to the exposure of the surgical field and the convenience of surgical operation, but also directly affects the patient's physiological function, postoperative recovery and even the safety of the operation. With the diversification of surgical types, the demand for surgical positions has become increasingly complex, such as the prone position in neurosurgery and the lateral position in cardiac surgery, which require medical staff to have superb position management skills and rich clinical experience. Therefore, ensuring the comfort and safety of patients during surgery has become one of the focuses of medical workers.

[0003] In one prior art, in order to ensure the smooth progress of the operation and improve the safety and efficiency of the operation, accurate adjustment of the patient's position is particularly important. After the patient is anesthetized, an experienced medical team will first safely move him to the operating table and place him in a roughly suitable starting position according to the type of operation to be performed. Next, in order to further accurately adjust the patient's position, the operating room is equipped with manipulators with multiple degrees of freedom joints. The design of these manipulators allows them to move flexibly in three-dimensional space, so that the patient's position can be carefully adjusted to meet the special needs of the surgical operation.

[0004] However, in the prior art, it is difficult for the robot to accurately detect pressure while moving the body, which results in excessive pressure during operation and the risk of causing secondary injuries. Summary of the invention

[0005] The present invention provides a method and system for adjusting the angle of a three-motor diagnosis and treatment bed based on intelligent body sensing, so as to avoid secondary injuries caused by moving patients during surgery.

[0006] In the first aspect, in order to solve the above technical problems, the present invention provides a method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing, comprising:

[0007] Obtain the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data;

[0008] Based on the pressure distribution data and the posture data, obtaining a first somatosensory feature by constructing a first equation;

[0009] Based on the pressure distribution data and the posture data, obtaining a second somatosensory feature through a second equation;

[0010] fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature;

[0011] The first somatosensory feature, the second somatosensory feature, and the third somatosensory feature are matched with the diagnosis and treatment demand data through a preset intelligent mapping model, and converted into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, to obtain mapping relationship data, wherein the initial deep learning model is trained according to the pre-stored historical somatosensory feature data and the historical diagnosis and treatment demand data, and the trained initial deep learning model is used as the intelligent mapping model;

[0012] An angle adjustment instruction is obtained according to the mapping relationship data conversion, and a driving motor is controlled according to the angle adjustment instruction to adjust the angle of the bed.

[0013] In an optional implementation, the obtaining of the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and the diagnosis and treatment demand data includes:

[0014] The pressure distribution data of various parts of the patient's body on the bed surface is obtained through the pressure sensor;

[0015] Acquire the patient's body posture data through a depth camera;

[0016] Acquire diagnosis and treatment demand data through user input operations.

[0017] In an optional implementation, obtaining the first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data includes:

[0018] Constructing a contact area distribution matrix and a pressure value matrix according to the pressure distribution data;

[0019] Extracting head position coordinates according to the posture data;

[0020] Constructing a pressure coefficient matrix according to the contact area distribution matrix and the pressure value matrix;

[0021] Constructing a pressure threshold matrix according to the posture data and the pressure value matrix;

[0022] The head position coordinates, the pressure coefficient matrix and the pressure threshold matrix are operated by constructing a first equation to obtain a first somatosensory feature;

[0023] The elements in the pressure threshold matrix are calculated by the following formula:

[0024]

[0025] The first somatosensory feature is calculated by the following formula:

[0026]

[0027] Among them, x1 represents the first somatosensory feature, F1 represents the first somatosensory feature, C ij Represents the elements in the pressure coefficient matrix, P ij Represents the elements in the pressure value matrix, A ij represents an element in the contact area distribution matrix, H represents the head position coordinate, and w1, w2 and w3 represent preset weight coefficients.

[0028] In an optional implementation, obtaining a second somatosensory feature through a second equation based on the pressure distribution data and the posture data includes:

[0029] The second somatosensory feature is obtained according to the following formula:

[0030]

[0031] Where x2 represents the second somatosensory feature, n represents the total amount of pressure distribution data and posture data; u i represents the ith pressure distribution data, v i represents the i-th posture data, represents the average value of the pressure distribution data, represents the average value of posture data; α represents the preset weight, and s represents the interval time for obtaining data of various parts of the patient's body.

[0032] In an optional implementation, the fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature includes:

[0033] Standardizing the first somatosensory feature and the second somatosensory feature to obtain standardized data;

[0034] Performing weighted fusion according to the standardized data to obtain a third somatosensory feature;

[0035] Among them, normalization is performed according to the following formula:

[0036]

[0037]

[0038] Among them, weighted fusion is performed according to the following formula:

[0039]

[0040] Among them, x′1 represents the standardized first feature, x′2 represents the standardized second feature, x1 represents the first somatosensory feature, x2 represents the second somatosensory feature, μ1 and μ2 are the means of the first somatosensory feature and the second somatosensory feature respectively, σ1 and σ2 are the standard deviations of the first somatosensory feature and the second somatosensory feature respectively, ω1 represents the preset first feature weight, ω2 represents the preset second feature weight, and x3 represents the third feature.

[0041] In an optional implementation, the training process of the initial deep learning model includes:

[0042] Inputting the pre-stored historical feature data and historical sample data into the input layer of the initial deep learning model for training, and obtaining the predicted value data output by the output layer of the deep learning model;

[0043] Substituting the historical body sensory feature data and the historical diagnosis and treatment demand data into a loss function to calculate a loss value, thereby obtaining loss value data;

[0044] Calculate the gradient of the output layer output of the deep learning model according to the loss value data, and pass the gradient forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer to obtain gradient data;

[0045] Update the parameters of each layer of the deep learning model based on the gradient data and the preset learning rate;

[0046] The parameters of each layer are updated repeatedly until the training is completed when the number of training times of the deep learning model is greater than the preset number of times, or when the loss value data of the deep learning model is less than the preset loss threshold.

[0047] In an optional implementation manner, converting the mapping relationship data to obtain the angle adjustment instruction includes:

[0048] Performing data analysis on the mapping relationship data to extract angle-affecting parameters;

[0049] Based on the angle influencing parameters, the bed adjustment angle is obtained by inverse kinematics calculation;

[0050] According to the bed adjustment angle, an angle adjustment instruction is obtained by performing calculations through a PID control algorithm;

[0051] The angle adjustment instruction includes controlling the first drive to adjust the inclination angle of the bed, controlling the second drive to adjust the pitch angle of the bed, or controlling the third drive to adjust the rotation angle of the bed.

[0052] In a second aspect, the present invention provides a three-motor diagnostic bed angle adjustment system based on intelligent body sensing, comprising:

[0053] A data acquisition module is used to obtain the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data;

[0054] A first feature acquisition module, configured to obtain a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data;

[0055] A second feature acquisition module, configured to acquire a second somatosensory feature through a second equation based on the pressure distribution data and the posture data;

[0056] A third feature acquisition module, used for fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature;

[0057] A data mapping module, used for matching the first somatosensory feature, the second somatosensory feature and the third somatosensory feature with the diagnosis and treatment demand data through a preset intelligent mapping model, converting them into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, and obtaining mapping relationship data, wherein an initial deep learning model is trained according to pre-stored historical somatosensory feature data and historical diagnosis and treatment demand data, and the trained initial deep learning model is used as an intelligent mapping model;

[0058] The bed adjustment module is used to obtain an angle adjustment instruction according to the mapping relationship data conversion, and control the drive motor to adjust the angle of the bed according to the angle adjustment instruction.

[0059] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned methods for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing.

[0060] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned three-motor diagnostic bed angle adjustment methods based on intelligent body sensing.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention relates to the field of intelligent control technology, and discloses a method and system for adjusting the angle of a three-motor medical bed based on intelligent somatosensory. The method includes obtaining pressure distribution data of various parts of a patient's body on a bed surface, posture data of the patient's body, and medical treatment demand data; obtaining a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data; obtaining a second somatosensory feature by a second equation based on the pressure distribution data and the posture data; fusing the first somatosensory feature with the second somatosensory feature to obtain a third somatosensory feature; matching the first somatosensory feature, the second somatosensory feature, and the third somatosensory feature with the medical treatment demand data through a preset intelligent mapping model, and converting them into a mapping relationship between somatosensory features and medical treatment demands to obtain mapping relationship data, wherein an initial deep learning model is trained according to pre-stored historical somatosensory feature data and historical medical treatment demand data, and the trained initial deep learning model is used as an intelligent mapping model; an angle adjustment instruction is converted according to the mapping relationship data, and a drive motor is controlled to adjust the angle of the bed according to the angle adjustment instruction. This method obtains the patient's pressure distribution, posture and treatment demand data, constructs and integrates somatosensory features, uses a trained intelligent mapping model to match somatosensory features with treatment needs, and finally converts them into angle adjustment instructions, thereby avoiding secondary injuries caused by moving the patient during the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic flow chart of a method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing provided by the first embodiment of the present invention;

[0064] Figure 2 It is a schematic diagram of the structure of a three-motor treatment bed angle adjustment system based on intelligent body sensing provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0065] 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.

[0066] Reference Figure 1 The first embodiment of the present invention provides a method for adjusting the angle of a three-motor diagnosis and treatment bed based on intelligent body sensing, comprising the following steps:

[0067] S11, obtaining pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data;

[0068] S12, obtaining a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data;

[0069] S13, obtaining a second somatosensory feature through a second equation based on the pressure distribution data and the posture data;

[0070] S14, fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature;

[0071] S15, matching the first somatosensory feature, the second somatosensory feature, and the third somatosensory feature with the diagnosis and treatment demand data through a preset intelligent mapping model, converting them into a mapping relationship between somatosensory features and diagnosis and treatment demands, and obtaining mapping relationship data, wherein an initial deep learning model is trained according to pre-stored historical somatosensory feature data and historical diagnosis and treatment demand data, and the trained initial deep learning model is used as an intelligent mapping model;

[0072] S16, obtaining an angle adjustment instruction according to the mapping relationship data conversion, and controlling the drive motor to adjust the angle of the bed according to the angle adjustment instruction.

[0073] In step S11, the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and the diagnosis and treatment demand data are obtained.

[0074] In a specific implementation, the obtaining of pressure distribution data of various parts of the patient's body on the bed surface, posture data of the patient's body, and diagnosis and treatment demand data includes:

[0075] The pressure distribution data of various parts of the patient's body on the bed surface is obtained through the pressure sensor;

[0076] Acquire the patient's body posture data through a depth camera;

[0077] Acquire diagnosis and treatment demand data through user input operations.

[0078] Specifically, first of all, through the pressure sensor array laid inside the mattress, such as smart skin electrical sensing gloves, smart skin electrical sensing vests, and smart pressure insoles, these sensors can accurately obtain the pressure distribution data of various parts of the patient's body on the bed surface, and sensitively respond to pressure changes in different parts of the human body. Whether it is slight pressure or greater pressure, it can be accurately captured and recorded.

[0079] Next, in order to obtain the patient's body posture data, a depth camera is used to capture the patient's three-dimensional body shape, and through image processing algorithms, key bone points of the human body, such as the head, shoulders, elbows, hip joints, knees and ankles, are accurately located. The acquisition of this data is used for subsequent posture analysis and pressure assessment and helps the system accurately determine the patient's sleeping posture and whether this sleeping posture will cause excessive pressure on certain parts of the body.

[0080] Finally, in order to better meet the personalized diagnosis and treatment needs of patients, the system also provides a user-friendly interface to obtain diagnosis and treatment demand data through user input operations. This part of the data includes the patient's medical history, current health status, and previous treatment experience.

[0081] In step S12, a first somatosensory feature is obtained by constructing a first equation based on the pressure distribution data and the posture data.

[0082] In a specific implementation, the obtaining the first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data includes:

[0083] Constructing a contact area distribution matrix and a pressure value matrix according to the pressure distribution data;

[0084] Extracting head position coordinates according to the posture data;

[0085] Constructing a pressure coefficient matrix according to the contact area distribution matrix and the pressure value matrix;

[0086] Constructing a pressure threshold matrix according to the posture data and the pressure value matrix;

[0087] The head position coordinates, the pressure coefficient matrix and the pressure threshold matrix are operated by constructing a first equation to obtain a first somatosensory feature;

[0088] The elements in the pressure threshold matrix are calculated by the following formula:

[0089]

[0090] The first somatosensory feature is calculated by the following formula:

[0091]

[0092] Among them, x1 represents the first somatosensory feature, C ij Represents the elements in the pressure coefficient matrix, P ij Represents the elements in the pressure value matrix, A ij represents an element in the contact area distribution matrix, H represents the head position coordinate, and w1, w2 and w3 represent preset weight coefficients.

[0093] Specifically, first, based on the pressure distribution data of various parts of the patient's body on the bed surface, two key matrices are constructed: the contact area distribution matrix A and the pressure value matrix P. These two matrices record the contact area size of each part of the patient's body with the bed surface and the corresponding pressure value. Each element A in the contact area distribution matrix A ij Represents the area of ​​contact between a specific area of ​​the patient's body and the bed surface; and each element P in the pressure value matrix P ij The corresponding pressure value of the area. These two matrices are the basis for subsequent calculations and can help analyze the stress conditions of various parts of the patient's body.

[0094] Assuming that m×n pressure sensors are arranged on the mattress, the contact area distribution matrix A can be expressed as:

[0095]

[0096] Among them, A ij It represents the contact area detected by the sensor at the position of row i and column j. These contact area values ​​are obtained by calibrating and processing the output signal of the sensor.

[0097] Assuming that i×j pressure sensors are arranged on the mattress, the pressure value matrix P can be expressed as:

[0098]

[0099] Among them, P ij It represents the pressure value detected by the sensor at the position of row i and column j. These pressure values ​​are usually obtained by calibrating and processing the output signal of the sensor.

[0100] Next, the head position coordinates H are extracted based on the patient's body posture data obtained by the depth camera. The determination of the head position is crucial to understanding the patient's overall posture and possible discomfort, because the relative position of the head can directly affect the pressure distribution of the neck and shoulders, and thus affect the patient's comfort.

[0101] After obtaining the above basic data, the pressure coefficient matrix C is further constructed based on the contact area distribution matrix A and the pressure value matrix P. Each element C in the pressure coefficient matrix ij Calculated by the following formula:

[0102]

[0103] Normalize pressure values ​​to more accurately reflect the amount of pressure per unit area.

[0104] Then, the pressure threshold matrix T is constructed according to the patient's body posture data and the pressure value matrix P. Each element T in the pressure threshold matrix ij This reflects the pressure threshold at which a part of the patient's body reaches a level of discomfort. This step requires setting a reasonable threshold based on the part of the body where the sensor is located to ensure that potential risk areas can be accurately identified.

[0105] Assuming that i×j pressure sensors are arranged on the mattress, the pressure threshold matrix T can be expressed as:

[0106]

[0107] Among them, T ij represents the pressure threshold detected by the sensor at the i-th row and j-th column. ij Set according to the part of the body where the sensor is located. For example, if sensor T 11 Located in the head area, then T 11 Can be set to 10 units of pressure; if sensor T 22 Located in the hip area, then T 22 The pressure can be set to 60 units. These values ​​are just examples and can be adjusted according to individual differences in actual applications.

[0108] Finally, the head position coordinates H, the pressure coefficient matrix C and the pressure threshold matrix T are used as inputs and substituted into the preset first formula for comprehensive calculation, thereby obtaining the first somatosensory feature x1. This is achieved through the following formula:

[0109]

[0110] Among them, x1 represents the first somatosensory feature, C ij Represents the elements in the pressure coefficient matrix, P ij Represents the elements in the pressure value matrix, A ij represents the elements in the contact area distribution matrix, H represents the head position coordinates, w1, w2 and w3 represent the preset weight coefficients, which are used to adjust the relative importance of each factor in the final somatosensory feature.

[0111] The addition of the head position coordinate H takes into account the impact of the head position on the overall body sensation; the sum of the pressure coefficient matrix C reflects the sum of the pressure per unit area of ​​each part of the patient's body; and the sum of the pressure threshold matrix T reflects the cumulative effect of the potential discomfort area. In this way, the system can comprehensively evaluate the patient's body sensation state.

[0112] In step S13, a second somatosensory feature is obtained through a second equation based on the pressure distribution data and the posture data.

[0113] In a specific implementation, the obtaining the second somatosensory feature through a second equation based on the pressure distribution data and the posture data includes:

[0114] The second somatosensory feature is obtained according to the following formula:

[0115]

[0116] Where x2 represents the second somatosensory feature, n represents the total amount of pressure distribution data and posture data; u i represents the ith pressure distribution data, v i represents the i-th posture data, represents the average value of the pressure distribution data, represents the average value of posture data; α represents the preset weight, and s represents the interval time for obtaining data of various parts of the patient's body.

[0117] Specifically, the average value of the pressure distribution data is calculated by the following formula:

[0118]

[0119] The average value of the posture data is calculated by the following formula:

[0120]

[0121] Next, calculate the correlation between the pressure distribution data and the posture data. This step is achieved by calculating the covariance of the two. The specific formula is as follows:

[0122]

[0123] This formula calculates the difference between the pressure distribution value of each data point and the average pressure distribution value, multiplies it by the difference between the posture value of each data point and the average posture value, and then sums all the data points. This result reflects the linear relationship between the pressure distribution data and the posture data.

[0124] In order to standardize the time scale and ensure the comparability of the calculation results, it is necessary to introduce the interval time for obtaining data.

[0125] Finally, the above calculation results are substituted into the formula of the second equation below to calculate the second somatosensory feature:

[0126]

[0127] Where x2 represents the second somatosensory feature, n represents the total amount of pressure distribution data and posture data; u i represents the ith pressure distribution data, v i represents the i-th posture data, represents the average value of the pressure distribution data, represents the average value of posture data; α represents the preset weight, and s represents the interval time for obtaining data of various parts of the patient's body.

[0128] Through the above steps, the second somatosensory feature is obtained. This feature value reflects the correlation between the pressure distribution data and the posture data, and can be used for further analysis and diagnosis.

[0129] In step S14, the first somatosensory feature and the second somatosensory feature are fused to obtain a third somatosensory feature.

[0130] In a specific implementation, the fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature includes:

[0131] Standardizing the first somatosensory feature and the second somatosensory feature to obtain standardized data;

[0132] Performing weighted fusion according to the standardized data to obtain a third somatosensory feature;

[0133] Among them, normalization is performed according to the following formula:

[0134]

[0135] Among them, weighted fusion is performed according to the following formula:

[0136]

[0137] Among them, x′1 represents the standardized first feature, x′2 represents the standardized second feature, x1 represents the first somatosensory feature, x2 represents the second somatosensory feature, μ1 and μ2 are the means of the first somatosensory feature and the second somatosensory feature respectively, σ1 and σ2 are the standard deviations of the first somatosensory feature and the second somatosensory feature respectively, ω1 represents the preset first feature weight, ω2 represents the preset second feature weight, and x3 represents the third feature.

[0138] Specifically, first, the first somatosensory feature and the second somatosensory feature are standardized to obtain standardized data. The purpose of the standardization process is to eliminate the dimensional differences between different features so that they can be compared and integrated on the same scale. The standardization formula is as follows:

[0139]

[0140]

[0141] Among them, x′1 represents the standardized first feature, x′2 represents the standardized second feature, x1 represents the first somatosensory feature, x2 represents the second somatosensory feature, v1 and μ2 are the means of the first somatosensory feature and the second somatosensory feature respectively, σ1 and σ2 are the standard deviations of the first somatosensory feature and the second somatosensory feature respectively.

[0142] Next, weighted fusion is performed on the standardized data to obtain the third somatosensory feature. The purpose of weighted fusion is to assign different weights to different features according to different application scenarios to highlight the importance of certain features. The weighted fusion formula is as follows:

[0143]

[0144] Among them, x′1 represents the standardized first feature, x′2 represents the standardized second feature, ω1 represents the preset first feature weight, ω2 represents the preset second feature weight, and x3 represents the third feature. The weights ω1 and ω2 can be adjusted according to specific application requirements to ensure that the final third somatosensory feature can better reflect the actual somatosensory state.

[0145] Through the above steps, a third somatosensory feature is obtained. This feature value combines the information of the first somatosensory feature and the second somatosensory feature, can more comprehensively reflect the patient's somatosensory state, and can more accurately evaluate the patient's body position changes and pressure distribution.

[0146] In step S15, the first somatosensory feature, the second somatosensory feature and the third somatosensory feature are matched with the diagnosis and treatment demand data through a preset intelligent mapping model, and converted into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, to obtain mapping relationship data, wherein the initial deep learning model is trained according to the pre-stored historical somatosensory feature data and the historical diagnosis and treatment demand data, and the trained initial deep learning model is used as the intelligent mapping model.

[0147] In a specific implementation, the training process of the initial deep learning model includes:

[0148] Inputting the pre-stored historical feature data and historical sample data into the input layer of the initial deep learning model for training, and obtaining the predicted value data output by the output layer of the deep learning model;

[0149] Substituting the historical body sensory feature data and the historical diagnosis and treatment demand data into a loss function to calculate a loss value, thereby obtaining loss value data;

[0150] Calculate the gradient of the output layer output of the deep learning model according to the loss value data, and pass the gradient forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer to obtain gradient data;

[0151] Update the parameters of each layer of the deep learning model based on the gradient data and the preset learning rate;

[0152] The parameters of each layer are updated repeatedly until the training is completed when the number of training times of the deep learning model is greater than the preset number of times, or when the loss value data of the deep learning model is less than the preset loss threshold.

[0153] Specifically, the new somatosensory feature data (first somatosensory feature, second somatosensory feature and third somatosensory feature) are input into the trained intelligent mapping model. The intelligent mapping model outputs the diagnosis and treatment demand data corresponding to the new somatosensory feature data through forward propagation calculation. The output data obtained is the mapping relationship data between the somatosensory feature and the diagnosis and treatment demand.

[0154] Through the above steps, the deep learning model can be effectively used to establish a mapping relationship between somatosensory characteristics and diagnosis and treatment needs, thereby improving the accuracy and efficiency of adjusting the angle.

[0155] In step S16, an angle adjustment instruction is obtained according to the mapping relationship data conversion, and the drive motor is controlled to adjust the angle of the bed according to the angle adjustment instruction.

[0156] In a specific implementation manner, the converting the mapping relationship data to obtain the angle adjustment instruction includes:

[0157] Performing data analysis on the mapping relationship data to extract angle-affecting parameters;

[0158] Based on the angle influencing parameters, the bed adjustment angle is obtained by inverse kinematics calculation;

[0159] According to the bed adjustment angle, an angle adjustment instruction is obtained by performing calculations through a PID control algorithm;

[0160] The angle adjustment instruction includes controlling the first drive to adjust the inclination angle of the bed, controlling the second drive to adjust the pitch angle of the bed, or controlling the third drive to adjust the rotation angle of the bed.

[0161] Specifically, the mapping relationship data is analyzed to extract the angle influencing parameters. The mapping relationship data contains the correspondence between the somatosensory characteristics and the diagnosis and treatment needs. These data include multiple parameters, such as pressure distribution, posture, and heart rate. By analyzing these data, parameters related to the bed angle adjustment can be extracted, such as the unevenness of pressure distribution and the pressure value of a specific part. Assume that the extracted angle influencing parameters are a1, a2, ..., a m , where m represents the number of parameters.

[0162] Based on the angle influencing parameters, the bed adjustment angle is obtained by inverse kinematics calculation. Inverse kinematics is an important concept in robotics, which is used to calculate the angles that each joint needs to rotate to achieve a specific position and posture. The inverse kinematics formula can be expressed as:

[0163] θ=f -1 (a1, a2, ..., a m )

[0164] Among them, θ represents the bed adjustment angle, f- 1 Represents the inverse kinematics function. The specific form of the inverse kinematics function depends on the mechanical structure and kinematic model of the bed.

[0165] According to the bed adjustment angle, the angle adjustment instruction is obtained by calculating through the PID control algorithm. The PID control algorithm is a commonly used feedback control algorithm that adjusts the behavior of the system through three parts: proportional, integral and differential. The formula of the PID control algorithm can be expressed as:

[0166]

[0167] Where u(t) represents the control signal, i.e., the angle adjustment instruction; K p , K i and K d are the gain coefficients of proportional, integral and differential respectively; e(τ) represents the error at the current moment, that is, the difference between the target angle and the current angle. For example, assuming the target adjustment angle is θ t , the current angle is θ c , then the error e(t) can be expressed as:

[0168] e(t)=θ t -θ c

[0169] According to the angle adjustment instruction calculated by the PID control algorithm, a specific control command is generated to control the drive motor to adjust the angle of the bed. The angle adjustment instruction may include controlling the first drive to adjust the inclination angle of the bed, controlling the second drive to adjust the pitch angle of the bed, or controlling the third drive to adjust the rotation angle of the bed.

[0170] The following describes the working process of the present invention using a relatively common scenario as an example. Figure 1 , a method for adjusting the angle of a three-motor diagnosis and treatment bed based on intelligent body sensing, comprising the following steps:

[0171] First, the pressure sensor array inside the mattress starts working. These sensors can accurately detect the pressure distribution of various parts of the patient's body on the bed surface. At the same time, the depth camera installed in the corner of the room starts, which uses advanced image processing technology to capture the patient's three-dimensional posture and accurately locate the key bone points of the human body, such as the head, shoulders, elbows, etc. In addition, medical staff input the patient's diagnosis and treatment needs, including medical history, current health status and other information through an intuitive user interface.

[0172] Subsequently, the system constructed a contact area distribution matrix and a pressure value matrix based on the collected pressure distribution data and posture data. Next, the head position coordinates were extracted from the data obtained by the depth camera. Using this information, the system constructed a pressure coefficient matrix and a pressure threshold matrix through a series of mathematical operations. Based on these matrices, the system calculated the first somatosensory feature, which reflects the force applied to various parts of the patient's body.

[0173] Next, the system uses the pressure distribution data and posture data again to calculate the covariance of the two to obtain the second somatosensory feature. This step further reveals the relationship between pressure distribution and patient posture. The calculation of the second somatosensory feature takes into account the time interval to ensure the accuracy of the result.

[0174] In order to comprehensively evaluate the patient's physical state, the system standardized the first and second physical characteristics, and weighted them according to the preset weights to finally obtain the third physical characteristic. This characteristic value not only takes into account the pressure distribution of various parts of the patient's body, but also takes into account the changes in posture, providing a scientific basis for subsequent bed adjustments.

[0175] Finally, the system inputs the first, second, and third somatosensory features together with the patient's diagnosis and treatment needs data into the pre-trained intelligent mapping model. Through this model, the system can accurately predict the best bed adjustment plan. Based on the results of the model output, the system uses inverse kinematics calculations and PID control algorithms to generate specific bed adjustment instructions. These instructions are sent to the three drive motors of the bed, which are responsible for adjusting the inclination, pitch, and rotation angles of the bed to achieve the most comfortable position adjustment.

[0176] As the bed was slowly adjusted to the optimal position, the patient suffered no secondary injury and the medical staff also obtained a more accurate surgical position.

[0177] In summary, the three-motor treatment bed angle adjustment system based on intelligent somatosensory accurately obtains the pressure distribution data and posture data of various parts of the patient's body through the pressure sensor array and depth camera laid inside the mattress. At the same time, the patient's personalized treatment demand data is collected through a user-friendly interface. Based on these data, the system constructs the contact area distribution matrix, pressure value matrix, pressure coefficient matrix and pressure threshold matrix, and calculates the first somatosensory feature and the second somatosensory feature. These two feature values ​​are standardized and weighted fused to generate the third somatosensory feature that comprehensively reflects the patient's somatosensory state. The system uses a pre-trained intelligent mapping model to match the first somatosensory feature, the second somatosensory feature and the third somatosensory feature with the patient's treatment demand data, and converts them into a mapping relationship between somatosensory features and treatment needs. Through this mapping relationship, the system can accurately predict the best bed adjustment plan. Finally, the system uses inverse kinematics calculation and PID control algorithm to generate specific bed adjustment instructions to control the drive motor to adjust the angle of the bed to achieve the most comfortable position adjustment.

[0178] The entire process not only reduces the operational burden on medical staff, but also avoids secondary injuries that may be caused by manual adjustment of the bed, significantly improving the safety and efficiency of surgery and diagnosis and treatment.

[0179] Reference Figure 2 The second embodiment of the present invention provides a three-motor diagnosis and treatment bed angle adjustment system based on intelligent body sensing, comprising:

[0180] A data acquisition module is used to obtain the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data;

[0181] A first feature acquisition module, configured to obtain a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data;

[0182] A second feature acquisition module, configured to acquire a second somatosensory feature through a second equation based on the pressure distribution data and the posture data;

[0183] A third feature acquisition module, used for fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature;

[0184] A data mapping module, used for matching the first somatosensory feature, the second somatosensory feature and the third somatosensory feature with the diagnosis and treatment demand data through a preset intelligent mapping model, converting them into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, and obtaining mapping relationship data, wherein an initial deep learning model is trained according to pre-stored historical somatosensory feature data and historical diagnosis and treatment demand data, and the trained initial deep learning model is used as an intelligent mapping model;

[0185] The bed adjustment module is used to obtain an angle adjustment instruction according to the mapping relationship data conversion, and control the drive motor to adjust the angle of the bed according to the angle adjustment instruction.

[0186] It should be noted that the three-motor medical bed angle adjustment device based on intelligent body sensing provided in an embodiment of the present invention is used to execute all the process steps of the three-motor medical bed angle adjustment method based on intelligent body sensing in the above-mentioned embodiment. The working principles and beneficial effects of the two correspond one to one, and therefore will not be repeated here.

[0187] The embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a three-motor diagnostic bed angle adjustment program based on intelligent body sensing. When the processor executes the computer program, the steps in the above-mentioned three-motor diagnostic bed angle adjustment method based on intelligent body sensing are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the angle adjustment module of the three-motor diagnosis and treatment bed based on intelligent body sensing.

[0188] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0189] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0190] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0191] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0192] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0193] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0194] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing, characterized in that: Executed by the controller, including: Obtain the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data; Based on the pressure distribution data and the posture data, obtaining a first somatosensory feature by constructing a first equation; Based on the pressure distribution data and the posture data, obtaining a second somatosensory feature through a second equation; fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature; The first somatosensory feature, the second somatosensory feature, and the third somatosensory feature are matched with the diagnosis and treatment demand data through a preset intelligent mapping model, and converted into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, to obtain mapping relationship data, wherein the initial deep learning model is trained according to the pre-stored historical somatosensory feature data and the historical diagnosis and treatment demand data, and the trained initial deep learning model is used as the intelligent mapping model; An angle adjustment instruction is obtained according to the mapping relationship data conversion, and a driving motor is controlled according to the angle adjustment instruction to adjust the angle of the bed.

2. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1 is characterized in that: The obtaining of the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and the diagnosis and treatment demand data includes: The pressure distribution data of various parts of the patient's body on the bed surface is obtained through the pressure sensor; Acquire the patient's body posture data through a depth camera; Acquire diagnosis and treatment demand data through user input operations.

3. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1 is characterized in that: The step of obtaining a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data includes: Constructing a contact area distribution matrix and a pressure value matrix according to the pressure distribution data; Extracting head position coordinates according to the posture data; Constructing a pressure coefficient matrix according to the contact area distribution matrix and the pressure value matrix; Constructing a pressure threshold matrix according to the posture data and the pressure value matrix; The head position coordinates, the pressure coefficient matrix and the pressure threshold matrix are operated by constructing a first equation to obtain a first somatosensory feature; The elements in the pressure threshold matrix are calculated by the following formula: The first somatosensory feature is calculated by the following formula: Among them, x1 represents the first somatosensory feature, C ij Represents the elements in the pressure coefficient matrix, P ij Represents the elements in the pressure value matrix, A ij represents an element in the contact area distribution matrix, H represents the head position coordinate, and w1, w2 and w3 represent preset weight coefficients.

4. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1, characterized in that: The obtaining a second somatosensory feature through a second equation based on the pressure distribution data and the posture data includes: The second somatosensory feature is obtained according to the following formula: Where x2 represents the second somatosensory feature, n represents the total amount of pressure distribution data and posture data; u i represents the ith pressure distribution data, v i represents the i-th posture data, represents the average value of the pressure distribution data, represents the average value of posture data; α represents the preset weight, and s represents the interval time for obtaining data of various parts of the patient's body.

5. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1, characterized in that: The fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature includes: Standardizing the first somatosensory feature and the second somatosensory feature to obtain standardized data; Performing weighted fusion according to the standardized data to obtain a third somatosensory feature; Among them, normalization is performed according to the following formula: Among them, weighted fusion is performed according to the following formula: Among them, x′1 represents the standardized first feature, x′2 represents the standardized second feature, x1 represents the first somatosensory feature, x2 represents the second somatosensory feature, μ1 and μ2 are the means of the first somatosensory feature and the second somatosensory feature respectively, σ1 and σ2 are the standard deviations of the first somatosensory feature and the second somatosensory feature respectively, ω1 represents the preset first feature weight, ω2 represents the preset second feature weight, and x3 represents the third feature.

6. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1, characterized in that: The training process of the initial deep learning model includes: Inputting the pre-stored historical feature data and historical sample data into the input layer of the initial deep learning model for training, and obtaining the predicted value data output by the output layer of the deep learning model; Substituting the historical body sensory feature data and the historical diagnosis and treatment demand data into a loss function to calculate a loss value, thereby obtaining loss value data; Calculate the gradient of the output layer output of the deep learning model according to the loss value data, and pass the gradient forward layer by layer through the chain rule to calculate the gradient of the parameters of each layer to obtain gradient data; Update the parameters of each layer of the deep learning model based on the gradient data and the preset learning rate; The parameters of each layer are updated repeatedly until the training is completed when the number of training times of the deep learning model is greater than the preset number of times, or when the loss value data of the deep learning model is less than the preset loss threshold.

7. The method for adjusting the angle of a three-motor diagnostic bed based on intelligent body sensing according to claim 1, characterized in that: The step of converting the mapping relationship data to obtain an angle adjustment instruction includes: Performing data analysis on the mapping relationship data to extract angle-affecting parameters; Based on the angle influencing parameters, the bed adjustment angle is obtained by inverse kinematics calculation; According to the bed adjustment angle, an angle adjustment instruction is obtained by performing calculations through a PID control algorithm; The angle adjustment instruction includes controlling the first drive to adjust the inclination angle of the bed, controlling the second drive to adjust the pitch angle of the bed, or controlling the third drive to adjust the rotation angle of the bed.

8. A three-motor diagnostic bed angle adjustment system based on intelligent body sensing, characterized in that: include: A data acquisition module is used to obtain the pressure distribution data of various parts of the patient's body on the bed surface, the patient's body posture data and diagnosis and treatment demand data; A first feature acquisition module, configured to obtain a first somatosensory feature by constructing a first equation based on the pressure distribution data and the posture data; A second feature acquisition module, configured to acquire a second somatosensory feature through a second equation based on the pressure distribution data and the posture data; A third feature acquisition module, used for fusing the first somatosensory feature and the second somatosensory feature to obtain a third somatosensory feature; A data mapping module, used for matching the first somatosensory feature, the second somatosensory feature and the third somatosensory feature with the diagnosis and treatment demand data through a preset intelligent mapping model, converting them into a mapping relationship between the somatosensory feature and the diagnosis and treatment demand, and obtaining mapping relationship data, wherein an initial deep learning model is trained according to pre-stored historical somatosensory feature data and historical diagnosis and treatment demand data, and the trained initial deep learning model is used as an intelligent mapping model; The bed adjustment module is used to obtain an angle adjustment instruction according to the mapping relationship data conversion, and control the drive motor to adjust the angle of the bed according to the angle adjustment instruction.

9. An electronic device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the angle adjustment method of the three-motor diagnostic bed based on intelligent body sensing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the angle adjustment method of a three-motor diagnostic bed based on intelligent body sensing as described in any one of claims 1 to 7.

Images