Intelligent monitoring method and system for realizing patient state based on vehicle-mounted sphygmomanometer

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent monitoring method and system for achieving the state of a patient based on a vehicle-mounted sphygmomanometer. Performing an infrared irradiation operation based on a preset initial infrared wave on the palm center position to obtain a palm center infrared reflection wave; vibration waves on a steering wheel are obtained, filtering operation is conducted on the palm center infrared reflection waves according to the vibration waves, noise reduction palm center reflection waves are obtained, difference calculation is conducted on the noise reduction palm center reflection waves and the initial infrared waves, and absorbed waves are obtained; utilizing a pre-trained variable pressure blood pressure recognition model to obtain a primary blood pressure recognition result according to the absorbed wave, and performing pressure correction operation on the primary blood pressure recognition result according to the steering wheel pressure matrix to obtain a blood pressure recognition result; and monitoring the blood pressure recognition result through a preset health threshold value. According to the invention, barrier-free blood pressure monitoring can be carried out on a driver, and the driving safety is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent monitoring method and system for a patient's status based on a vehicle-mounted sphygmomanometer. Background Art

[0002] The continuous improvement of social industrial level has made cars gradually popular among the public. The increase in vehicles has led to an increase in traffic accidents. With the enrichment of accident investigation data, people have found that the proportion of accidents caused by driver operation errors is much larger than that caused by vehicle failures. A large factor in driver operation errors is physical problems that cause the driver to lose concentration or control himself.

[0003] Due to the long-term sitting working environment, today, hypertensive patients can be seen everywhere among drivers. The physical health of drivers ensures the life safety of drivers, passengers and passers-by. Nowadays, there are also some car-mounted blood pressure monitors. These car-mounted blood pressure monitors are connected to the car's central control, but they still require patients to operate manually, such as manually tying air bags, pressing test buttons, etc., which not only wastes patients' time but also interferes with driving. How to effectively monitor the physical health of drivers without interfering with driving has become a problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides an intelligent monitoring method for a patient's status based on a vehicle-mounted sphygmomanometer, the main purpose of which is to perform barrier-free blood pressure monitoring on a driver and improve driving safety.

[0005] To achieve the above object, the present invention provides an intelligent monitoring method for patient status based on a vehicle-mounted sphygmomanometer, comprising: Using a pre-built pressure sensor, a steering wheel pressure matrix is ​​obtained, and according to the steering wheel pressure matrix, the palm position of the patient is identified to obtain palm position information; Using a pre-built infrared sensor, according to the palm position information, an infrared irradiation operation based on a preset initial infrared wave is performed on the palm position, and a reflection signal of the infrared irradiation operation is received to obtain a palm infrared reflection wave; Using a pre-built vibration sensor, a vibration wave on a pre-built steering wheel is obtained, filtering the palm infrared reflection wave according to the vibration wave to obtain a de-noised palm reflection wave, and performing a difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain an absorbed wave; Using a pre-trained variable pressure blood pressure recognition model, performing time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, performing a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and performing a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution; Performing a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; Performing a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and performing a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; Obtain a health threshold, and determine whether the blood pressure recognition result is greater than the health threshold; When the blood pressure recognition result is greater than the health threshold, the patient is determined to be driving dangerously and an alarm prompt message is generated.

[0006] Optionally, before acquiring the steering wheel pressure matrix using the pre-built pressure sensor, the method further includes: Get the patient's real-time location using pre-built in-car navigation; Obtaining the patient's historical common route, and determining whether the patient is in a straight driving section according to the historical common route and the real-time location; When the patient is in a straight driving section, it is determined that the patient is in a stable driving stage; When the patient is not in a straight driving section, it is determined that the patient is in an unstable driving stage; If the patient is in a stable driving stage, the operation of using the pre-built pressure sensor to obtain the steering wheel pressure matrix is ​​performed.

[0007] Optionally, identifying the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information includes: Performing a masking operation of a preset pressure on the steering wheel pressure matrix to obtain a pressure masking result; Performing an outermost contour edge recognition operation on the pressure masking result to obtain a hand shape distribution of the user; Performing a key node recognition operation on the hand shape distribution to obtain a key node set; The palm position is predicted according to the relative position of each key node in the key node set to obtain the palm position information.

[0008] Optionally, filtering the palm infrared reflected wave according to the vibration wave to obtain the de-noised palm reflected wave includes: Using a pre-built high-pass filter to perform DC component filtering on the palm infrared reflected wave to obtain a DC component-free wave; The DC component-removed wave is subjected to standardization processing based on a mean of zero and a variance of one to obtain an infrared standard wave; Using a pre-built fast Fourier transform algorithm, the vibration component of the vibration wave is identified to obtain a main vibration frequency; The infrared standard wave is filtered according to the main vibration frequency to obtain a de-noised palm reflection wave.

[0009] Optionally, the variable pressure blood pressure recognition model is expressed as:

[0010]

[0011] In the formula, represents the blood pressure recognition result, represents the primary blood pressure recognition result, represents the initial weight of the primary blood pressure recognition result for the blood pressure recognition result, represents the steering wheel pressure matrix, Represents the steering wheel pressure matrix Zhong The weight coefficient at the point position, Represents the steering wheel pressure matrix The horizontal axis in Represents the steering wheel pressure matrix The vertical coordinate in Indicates the bias of blood pressure recognition results, represents the bias of the primary blood pressure recognition result, Represents the waveform characteristics, represents the frequency characteristics, Represents the amplitude characteristics, , and All of these are characteristics of the red blood cell flow distribution. , and Represents the weight coefficient.

[0012] Optionally, before using the pre-trained variable pressure blood pressure recognition model, the method further includes: Obtaining a pre-built variable pressure blood pressure recognition model and obtaining a real blood pressure measurement value sample, wherein the real blood pressure measurement value sample includes a measured hand pressure distribution, a hand blood pressure value label and a real blood pressure value label; Sequentially acquiring a target sample from the real blood pressure measurement value samples, performing hand blood pressure prediction on the target sample using the variable pressure blood pressure recognition model to obtain a hand blood pressure prediction result, and performing real blood pressure prediction on the hand blood pressure result according to the measured hand pressure distribution in the target sample to obtain a real blood pressure prediction result; Using a pre-built cross entropy loss algorithm, the loss value between the hand blood pressure value label of the target sample and the hand blood pressure prediction result is calculated to obtain a first loss value, and the loss value between the true blood pressure value label of the target sample and the true blood pressure prediction result is calculated to obtain a second loss value; Performing weighted calculation on the first loss value and the second loss value to obtain a comprehensive loss value; According to the gradient descent algorithm, the comprehensive loss value is minimized to obtain the network model parameters when the comprehensive loss value is the minimum, and the network model parameters are subjected to a reverse parameter update operation to obtain an updated variable pressure blood pressure recognition model; Return to the above step of obtaining a target sample from the real blood pressure measurement value samples in sequence, and iteratively train the updated variable-pressure blood pressure recognition model. When the sample traversal in the updated variable-pressure blood pressure recognition model is completed, a trained variable-pressure blood pressure recognition model is obtained.

[0013] Optionally, performing time-frequency analysis on the absorbed wave to obtain a set of absorbed wave features includes: Performing a waveform feature extraction operation on the absorbed wave to obtain a time domain signal feature set; Using a pre-built short-time Fourier transform algorithm, the absorbed wave is subjected to a frequency domain conversion operation to obtain a frequency domain signal, and waveform feature extraction is performed on the frequency domain signal to obtain a frequency domain signal feature set; The time domain signal feature set and the frequency domain signal feature set are output to obtain an absorbed wave feature set.

[0014] Optionally, after obtaining the blood pressure recognition result, the method further includes: Recording the blood pressure recognition result to obtain a blood pressure change curve; Performing differential calculation on the blood pressure change curve to obtain a blood pressure change rate curve, and monitoring the blood pressure change rate curve according to a preset warning threshold; When the blood pressure change in the blood pressure change rate curve is greater than the warning threshold, the patient is determined to be driving dangerously, and the vehicle is automatically braked according to a preset emergency braking strategy.

[0015] Optionally, obtaining the health threshold includes: Use the pre-built vehicle-mounted central control device to obtain the patient's identity information; Querying a specific information set of the patient according to the identity information, wherein the specific information set includes age, complications and treatment history; According to the specific information set, a preset default hypertension threshold is weightedly calculated to obtain a healthy threshold.

[0016] To achieve the above object, the present invention also provides an intelligent monitoring system for patient status based on a vehicle-mounted sphygmomanometer, comprising: The infrared monitoring module is used to obtain a steering wheel pressure matrix using a pre-built pressure sensor, and identify the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information, and use the pre-built infrared sensor to perform an infrared irradiation operation based on a preset initial infrared wave on the palm position according to the palm position information, and receive a reflection signal of the infrared irradiation operation to obtain a palm infrared reflection wave; The blood absorption wave identification module is used to obtain the vibration wave on the pre-constructed steering wheel by using the pre-constructed vibration sensor, filter the palm infrared reflection wave according to the vibration wave to obtain the de-noised palm reflection wave, and perform difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain the absorbed wave; A palm blood pressure recognition module is used to use a pre-trained variable pressure blood pressure recognition model to perform time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, and to perform a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and to perform a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution, and to perform a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; a blood pressure correction module, configured to perform a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and perform a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; The supervision and warning module is used to obtain a health threshold, determine whether the blood pressure recognition result is greater than the health threshold, and when the blood pressure recognition result is greater than the health threshold, determine that the patient is driving dangerously and generate an alarm prompt message.

[0017] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising: a memory storing at least one instruction; and The processor executes the instructions stored in the memory to implement the above-mentioned intelligent monitoring method of the patient's status based on the vehicle-mounted blood pressure monitor.

[0018] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned intelligent monitoring method of patient status based on a vehicle-mounted blood pressure monitor.

[0019] The present invention is to solve the problem described in the background technology. The present invention first obtains the blood flow pattern of the patient's palm part through infrared scanning technology, thereby identifying the primary blood pressure recognition result of the patient's palm position, and then obtains the patient's grip on the steering wheel during driving to obtain the steering wheel pressure matrix. The present invention uses a pre-trained variable pressure blood pressure recognition model to explore the relationship between the grip of the human hand and the real blood pressure, thereby predicting the real blood pressure based on the steering wheel pressure matrix and the primary blood pressure recognition result to obtain the blood pressure recognition result. Among them, the sensor equipment in the present invention is configured in the steering wheel, and the data processing process provides computing power resources through the car central control to achieve no interference with the patient's driving process. Therefore, the present invention can perform barrier-free blood pressure monitoring on the driver and improve driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic flow chart of a method for intelligently monitoring a patient's status based on a vehicle-mounted sphygmomanometer provided in one embodiment of the present invention; Figure 2 A functional module diagram of an intelligent monitoring system for patient status based on a vehicle-mounted sphygmomanometer provided in one embodiment of the present invention; Figure 3 A schematic diagram of the structure of an electronic device for implementing the method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer provided in one embodiment of the present invention.

[0021] Description of figure markings: 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0022] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0024] The embodiment of the present application provides an intelligent monitoring method for patient status based on a vehicle-mounted sphygmomanometer. The execution subject of the intelligent monitoring method for patient status based on a vehicle-mounted sphygmomanometer includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the intelligent monitoring method for patient status based on a vehicle-mounted sphygmomanometer can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0025] Reference Figure 1 FIG. 1 is a flow chart of a method for intelligently monitoring a patient's status based on a vehicle-mounted blood pressure monitor according to an embodiment of the present invention. In this embodiment, the method for intelligently monitoring a patient's status based on a vehicle-mounted blood pressure monitor includes: S1. Using a pre-built pressure sensor, obtain a steering wheel pressure matrix, and identify the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information.

[0026] In the embodiment of the present invention, the pressure sensor may be placed on the outside of the steering wheel in the form of a film-type sensor, or may be placed inside the steering wheel in the form of a built-in sensor.

[0027] The patient is a patient with hypertension, and the patient in the embodiment plays the role of a driver.

[0028] Specifically, in the embodiment of the present invention, the pressure distribution of the palm of a hypertensive patient on the steering wheel is collected by a film-type sensor to obtain a steering wheel pressure matrix.

[0029] It should be known that when measuring blood pressure, patients are usually required to relax their hands, but when driving, they will hold the steering wheel tightly, which makes the hand blood pressure different from the normal blood pressure of the body. Therefore, collecting the steering wheel pressure matrix can not only know the position of the patient's hand, but also facilitate the study of the relationship between hand blood pressure and normal blood pressure.

[0030] In detail, in the embodiment of the present invention, before obtaining the steering wheel pressure matrix using the pre-built pressure sensor, the method further includes: Get the patient's real-time location using pre-built in-car navigation; Obtaining the patient's historical common route, and determining whether the patient is in a straight driving section according to the historical common route and the real-time location; When the patient is in a straight driving section, it is determined that the patient is in a stable driving stage; When the patient is not in a straight driving section, it is determined that the patient is in an unstable driving stage; If the patient is in a stable driving stage, the operation of using the pre-built pressure sensor to obtain the steering wheel pressure matrix is ​​performed.

[0031] The vehicle navigation system can obtain the real-time location of the vehicle driven by the patient based on a satellite positioning system or an offline map.

[0032] The historical commonly used route is a route that the patient often passes by, such as a commute route. In addition, when the patient walks to a new route, the navigation route can be used as the historical commonly used data.

[0033] Furthermore, the straight driving section refers to a section where the vehicle can move forward stably, and the reliance on steering wheel control is relatively reduced; The stable driving stage means that during the straight driving section, even if it is necessary to change lanes, the patient does not need to take his hands off the steering wheel to achieve control.

[0034] According to test statistics, during the stable driving stage, it is more difficult for the patient to concentrate, and artificial intelligence intervention monitoring is more needed. At this time, the patient's hands are stable on the steering wheel, making it easier to perform infrared scanning. Therefore, in the embodiment of the present invention, a monitoring start strategy is configured. If the patient is in the stable driving stage, the operation of using the pre-built pressure sensor to obtain the steering wheel pressure matrix is ​​executed.

[0035] In detail, in the embodiment of the present invention, the palm position of the patient is identified according to the steering wheel pressure matrix to obtain the palm position information, including: Performing a masking operation of a preset pressure on the steering wheel pressure matrix to obtain a pressure masking result; Performing an outermost contour edge recognition operation on the pressure masking result to obtain a hand shape distribution of the user; Performing a key node recognition operation on the hand shape distribution to obtain a key node set; The palm position is predicted according to the relative position of each key node in the key node set to obtain the palm position information.

[0036] The mask refers to an operation of selecting or excluding a specific part.

[0037] In the embodiment of the present invention, the values ​​in the steering wheel pressure matrix that are less than 500Pa are masked by masking operation, and the main force-bearing shape is retained, and generally the image of the palm appears. Then, the present invention performs the outermost contour edge recognition through the edge segmentation algorithm to obtain the hand shape distribution, and then performs the key node recognition operation, wherein the key nodes generally include the positions of five fingers and the base of the thumb. When three or more key nodes are obtained, the palm position information can be inferred based on the positional relationship between the key nodes, for example, the palm position is above the 10 o'clock position of the steering wheel.

[0038] Among them, the edge segmentation algorithm refers to an algorithm used to detect edges in an image and segment different objects or areas. It usually represents the boundary or structure of an object based on places in the image where the brightness and color changes greatly. The present invention uses the values ​​in the steering wheel pressure matrix as brightness for segmentation, and can easily obtain the hand shape distribution.

[0039] S2. Using a pre-built infrared sensor, according to the palm position information, perform an infrared irradiation operation based on a preset initial infrared wave on the palm position, and receive a reflection signal of the infrared irradiation operation to obtain a palm infrared reflection wave.

[0040] In the embodiment of the present invention, the infrared sensor has the function of emitting and receiving near infrared light. In addition, the infrared sensor is installed in the steering wheel and can be moved in the steering wheel by a motor or magnetic force.

[0041] Specifically, in the embodiment of the present invention, according to the palm position being above the 10 o'clock position of the steering wheel, the infrared sensor is moved to the palm position by motor pulling or electromagnetic track and then infrared light scanning is performed. In the embodiment of the present invention, near-infrared light with a wavelength of 700nm is used as the initial infrared wave for irradiation, and then after being absorbed by human bones, muscles, blood vessels and blood, the remaining part is returned to the infrared sensor to obtain the palm infrared reflection wave.

[0042] S3. Using a pre-built vibration sensor, obtain a vibration wave on a pre-built steering wheel, perform a filtering operation on the palm infrared reflection wave according to the vibration wave to obtain a de-noised palm reflection wave, and perform a difference calculation on the de-noised palm reflection wave and the initial infrared wave to obtain an absorbed wave.

[0043] The vibration sensor is a device that detects and measures the vibration or movement of an object. In the embodiment of the present invention, the vibration sensor is also installed in the steering wheel and is close to the infrared sensor.

[0044] In the embodiment of the present invention, the infrared sensor and the palm are both on the steering wheel. The vibration signal during vehicle driving will interfere with the collected palm infrared reflection wave. Therefore, a vibration sensor is needed to collect the vibration wave and eliminate the part of the signal involving vibration in the palm infrared reflection wave.

[0045] Specifically, in an embodiment of the present invention, filtering the palm infrared reflected wave according to the vibration wave to obtain the de-noised palm reflected wave includes: Using a pre-built high-pass filter to perform DC component filtering on the palm infrared reflected wave to obtain a DC component-free wave; The DC component-removed wave is subjected to standardization processing based on a mean of zero and a variance of one to obtain an infrared standard wave; Using a pre-built fast Fourier transform algorithm, the vibration component of the vibration wave is identified to obtain a main vibration frequency; The infrared standard wave is filtered according to the main vibration frequency to obtain a de-noised palm reflection wave.

[0046] The high-pass filter (HPF) is an electronic filter that allows high-frequency signals to pass through and suppresses low-frequency signals. In the embodiment of the present invention, the high-pass filter is used to remove the DC component in the signal so as to better observe the AC component.

[0047] The Fast Fourier Transform (FFT) algorithm is an algorithm for efficiently calculating the Discrete Fourier Transform (DFT) and its inverse transform. The Discrete Fourier Transform is a mathematical tool for converting signals from the time domain to the frequency domain, and is widely used in signal processing, image processing, audio analysis, communication systems, and other fields. The FFT significantly improves the computational efficiency by decomposing the DFT calculation into multiple smaller DFT calculations.

[0048] Specifically, in an embodiment of the present invention, the DC component is first filtered through a high-pass filter to obtain a DC component-removed wave, and then the DC component-removed wave is processed through a standardized normal distribution to obtain an infrared standard wave. The infrared standard wave can make different features have the same scale, thereby improving the effect and convergence speed of the algorithm when performing machine learning or statistical analysis.

[0049] Then, the present invention transforms the vibration wave of the time domain signal into a frequency domain signal through a fast Fourier transform algorithm, and then identifies the main peak components in the frequency domain signal to obtain the main vibration frequency. Finally, the signal about the active vibration frequency in the infrared standard wave is accurately filtered to obtain the noise-reduced palm reflection wave.

[0050] Furthermore, in an embodiment of the present invention, the absorbed wave is obtained by calculating the difference between the noise-reduced palm reflected wave and the initial infrared wave, wherein the absorbed wave is the wave absorbed by the muscles, bones, blood vessels and blood of the hand.

[0051] S4. Using a pre-trained variable-pressure blood pressure recognition model, perform time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, perform a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and perform a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution.

[0052] The variable pressure blood pressure recognition model is a neural network model, which is used to predict the real patient blood pressure (blood pressure recognition result) according to the hand pressure (steering wheel pressure matrix) and the hand blood pressure (primary blood pressure recognition result). The variable pressure blood pressure recognition model includes a blood pressure recognition network and a blood pressure correction network; the blood pressure recognition network recognizes blood pressure according to the distribution and flow rate of hemoglobin in the blood vessels, and the blood pressure correction network corrects the hand blood pressure according to the pressure at different positions of the hand to obtain the blood pressure measured under normal conditions.

[0053] In detail, in the embodiment of the present invention, the variable pressure blood pressure recognition model is expressed as:

[0054]

[0055] In the formula, represents the blood pressure recognition result, represents the primary blood pressure recognition result, represents the initial weight of the primary blood pressure recognition result for the blood pressure recognition result, represents the steering wheel pressure matrix, Represents the steering wheel pressure matrix Zhong The weight coefficient at the point position, Represents the steering wheel pressure matrix The horizontal axis in Represents the steering wheel pressure matrix The vertical coordinate in Indicates the bias of blood pressure recognition results, represents the bias of the primary blood pressure recognition result, Represents the waveform characteristics, represents the frequency characteristics, Represents the amplitude characteristics, , and All of these are characteristics of the red blood cell flow distribution. , and Represents the weight coefficient.

[0056] Specifically, in the embodiment of the present invention, the red blood cell flow distribution is first feature extracted to obtain features such as waveform, frequency and amplitude, and then the network parameter relationship pre-trained in the variable pressure blood pressure recognition model is used to establish the relationship between the red blood cell flow characteristics and the blood pressure, thereby achieving full connection calculation of features such as waveform, frequency and amplitude to obtain a primary blood pressure recognition result, and then, the steering wheel pressure matrix is ​​obtained, and the blood pressure value of the primary blood pressure recognition result is corrected through the regression relationship in the pre-trained network to obtain the final blood pressure recognition result.

[0057] In detail, in the embodiment of the present invention, before using the pre-trained variable pressure blood pressure recognition model, the method further includes: Obtaining a pre-built variable pressure blood pressure recognition model and obtaining a real blood pressure measurement value sample, wherein the real blood pressure measurement value sample includes a measured hand pressure distribution, a hand blood pressure value label and a real blood pressure value label; Sequentially acquiring a target sample from the real blood pressure measurement value samples, performing hand blood pressure prediction on the target sample using the variable pressure blood pressure recognition model to obtain a hand blood pressure prediction result, and performing real blood pressure prediction on the hand blood pressure result according to the measured hand pressure distribution in the target sample to obtain a real blood pressure prediction result; Using a pre-built cross entropy loss algorithm, the loss value between the hand blood pressure value label of the target sample and the hand blood pressure prediction result is calculated to obtain a first loss value, and the loss value between the true blood pressure value label of the target sample and the true blood pressure prediction result is calculated to obtain a second loss value; Performing weighted calculation on the first loss value and the second loss value to obtain a comprehensive loss value; According to the gradient descent algorithm, the comprehensive loss value is minimized to obtain the network model parameters when the comprehensive loss value is the minimum, and the network model parameters are subjected to a reverse parameter update operation to obtain an updated variable pressure blood pressure recognition model; Return to the above step of obtaining a target sample from the real blood pressure measurement value samples in sequence, and iteratively train the updated variable-pressure blood pressure recognition model. When the sample traversal in the updated variable-pressure blood pressure recognition model is completed, a trained variable-pressure blood pressure recognition model is obtained.

[0058] Among them, the real blood pressure measurement value sample refers to a set of values ​​of blood pressure measurement by the test person through various grip strengths and grip postures. The content of the sample in the real blood pressure measurement value sample is the measured hand pressure distribution, and the hand blood pressure value label and the real blood pressure value label are marks.

[0059] In the embodiment of the present invention, during the forward network calculation process of the variable-pressure blood pressure recognition model, the hand blood pressure value is first predicted to obtain a hand blood pressure prediction result, and then the hand blood pressure prediction result is used as input to predict the real blood pressure together with the measured hand pressure distribution of the target sample to obtain the real blood pressure prediction result.

[0060] In the embodiment of the present invention, both parts of the network are regression predictions, so the same cross entropy loss algorithm can be used to calculate the loss value, and finally a comprehensive loss value can be calculated, and the overall model training can be realized through the comprehensive loss value. The present invention uses a gradient descent algorithm to control the direction of model training, thereby obtaining an updated variable pressure blood pressure recognition model. Among them, the cross entropy loss algorithm is a method of measuring the difference between two probability distributions, and is often used to evaluate the similarity between the probability distribution of the model output and the true label distribution. The gradient descent algorithm is an iterative algorithm for optimizing model parameters, which is widely used in machine learning and deep learning. The main goal is to minimize the loss function so that the model can better fit the training data.

[0061] In the embodiment of the present invention, it is convenient to obtain the real blood pressure measurement value samples. After a real blood pressure measurement value sample traversal is completed, if the output result of the model still does not meet the standard, a new real blood pressure measurement value sample can be obtained for re-training.

[0062] After the trained variable-pressure blood pressure recognition model is obtained, the feature set of the absorbed wave is obtained by performing feature extraction calculation on the absorbed wave.

[0063] In detail, in the embodiment of the present invention, the time-frequency analysis of the absorbed wave to obtain the absorbed wave feature set includes: Performing a waveform feature extraction operation on the absorbed wave to obtain a time domain signal feature set; Using a pre-built short-time Fourier transform algorithm, the absorbed wave is subjected to a frequency domain conversion operation to obtain a frequency domain signal, and waveform feature extraction is performed on the frequency domain signal to obtain a frequency domain signal feature set; The time domain signal feature set and the frequency domain signal feature set are output to obtain an absorbed wave feature set.

[0064] The Short-Time Fourier Transform (STFT) algorithm is a tool for analyzing the frequency components of non-stationary signals. It combines the advantages of time and frequency analysis and can be used for time-frequency analysis of signals.

[0065] Specifically, in the embodiment of the present invention, a waveform feature extraction operation is first directly performed on the absorbed wave, and then the time domain signal is converted into a frequency domain signal, and then waveform feature recognition is performed again, and finally the two recognition results are comprehensively output to obtain a set of absorbed wave features.

[0066] Then continue the network process to calculate the red blood cell absorption wave feature set and red blood cell flow distribution. This is the forward operation process of the neural network model. Its specific structure is various and flexible, so it will not be described here.

[0067] S5. Perform a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result.

[0068] Specifically, in the embodiment of the present invention, through the hand blood pressure recognition operation, the red blood cell flow distribution is first convolved, pooled and flattened to extract features, and then the weight parameters obtained according to the model training are fully connected to obtain the primary blood pressure recognition result.

[0069] S6. Perform a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and perform a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result.

[0070] Specifically, in the embodiment of the present invention, similar to the steps in S5 above, this time the steering wheel pressure matrix and the primary blood pressure recognition results are used as input, and convolution, pooling and other operations are re-performed. Through the network model parameters corrected during the training process, full connection prediction is performed on each feature to obtain the blood pressure recognition result.

[0071] In detail, in the embodiment of the present invention, after obtaining the blood pressure recognition result, the method further includes: Recording the blood pressure recognition result to obtain a blood pressure change curve; Performing differential calculation on the blood pressure change curve to obtain a blood pressure change rate curve, and monitoring the blood pressure change rate curve according to a preset warning threshold; When the blood pressure change in the blood pressure change rate curve is greater than the warning threshold, the patient is determined to be driving dangerously, and the vehicle is automatically braked according to a preset emergency braking strategy.

[0072] In the embodiment of the present invention, when the blood pressure is higher than a certain threshold, such as 160 mmHg, dizziness, fainting, etc. may occur. However, if the blood pressure starts out normal and then suddenly increases, this situation is more dangerous and may be an acute hypertensive crisis, cardiac event, stress response, etc.

[0073] In an embodiment of the present invention, the blood pressure change curve is converted into a blood pressure change rate curve through a differential operation, and then a warning threshold is configured, for example, 20 mmHg per second. When the blood pressure change in the blood pressure change rate curve is greater than the warning threshold, this situation requires emergency avoidance, and the car can be pulled over through the car's central control or even automatically alarmed.

[0074] S7. Obtain a health threshold, and determine whether the blood pressure recognition result is greater than the health threshold.

[0075] In detail, in the embodiment of the present invention, obtaining the health threshold includes: Use the pre-built vehicle-mounted central control device to obtain the patient's identity information; Querying a specific information set of the patient according to the identity information, wherein the specific information set includes age, complications and treatment history; According to the specific information set, a preset default hypertension threshold is weightedly calculated to obtain a healthy threshold.

[0076] It should be known that different people have different physiques, and the degree of influence on blood pressure changes according to age changes is different. In the embodiment of the present invention, the influence of age, complications and treatment history on hypertension resistance is collected through big data, so as to fine-tune the patient's hypertension threshold. When the default hypertension threshold is 160 mmHg, some people may be configured to 150 or 170 mmHg.

[0077] The present invention can monitor the patient's condition by comparing whether the blood pressure identification result is greater than a healthy threshold.

[0078] S8. When the blood pressure recognition result is greater than the health threshold, the patient is determined to be driving dangerously and an alarm prompt message is generated.

[0079] The determination process of the blood pressure change rate curve in the embodiment of the present invention is similar to that in the above-mentioned embodiment. In the embodiment of the present invention, when the user's blood pressure slowly increases due to poor blood circulation caused by long-term driving and exceeds the health threshold, an alarm prompt message can be generated, such as a text such as "pay attention to rest" to improve the patient's driving vigilance.

[0080] The present invention is to solve the problem described in the background technology. The present invention first obtains the blood flow pattern of the patient's palm part through infrared scanning technology, thereby identifying the primary blood pressure recognition result of the patient's palm position, and then obtains the patient's grip on the steering wheel during driving to obtain the steering wheel pressure matrix. The present invention uses a pre-trained variable pressure blood pressure recognition model to explore the relationship between the grip of the human hand and the real blood pressure, thereby predicting the real blood pressure based on the steering wheel pressure matrix and the primary blood pressure recognition result to obtain the blood pressure recognition result. Among them, the sensor equipment in the present invention is configured in the steering wheel, and the data processing process provides computing power resources through the car central control to achieve no interference with the patient's driving process. Therefore, the present invention can perform barrier-free blood pressure monitoring on the driver and improve driving safety.

[0081] like Figure 2 , which is a functional module diagram of an intelligent monitoring system for patient status based on a vehicle-mounted sphygmomanometer provided by an embodiment of the present invention.

[0082] The intelligent monitoring system 100 for patient status based on the vehicle-mounted sphygmomanometer of the present invention can be installed in an electronic device. According to the functions implemented, the intelligent monitoring system 100 for patient status based on the vehicle-mounted sphygmomanometer can include an infrared monitoring module 101, a blood absorption wave recognition module 102, a palm blood pressure recognition module 103, a blood pressure correction module 104 and a supervision warning module 105. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, which are stored in the memory of the electronic device.

[0083] The infrared monitoring module 101 is used to obtain a steering wheel pressure matrix using a pre-built pressure sensor, and identify the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information, and use the pre-built infrared sensor to perform an infrared irradiation operation based on a preset initial infrared wave on the palm position according to the palm position information, and receive a reflection signal of the infrared irradiation operation to obtain a palm infrared reflection wave; The blood absorption wave identification module 102 is used to obtain the vibration wave on the pre-constructed steering wheel by using the pre-constructed vibration sensor, filter the palm infrared reflection wave according to the vibration wave to obtain the de-noised palm reflection wave, and perform difference calculation on the de-noised palm reflection wave and the initial infrared wave to obtain the absorbed wave; The palm blood pressure recognition module 103 is used to perform time-frequency analysis on the absorbed wave using a pre-trained variable pressure blood pressure recognition model to obtain an absorbed wave feature set, and to perform a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and to perform a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution, and to perform a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; The blood pressure correction module 104 is used to perform a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and perform a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; The supervision warning module 105 is used to obtain a health threshold, determine whether the blood pressure recognition result is greater than the health threshold, and when the blood pressure recognition result is greater than the health threshold, determine that the patient is driving dangerously and generate an alarm prompt message.

[0084] In detail, the modules in the intelligent monitoring system 100 for realizing the patient status based on the vehicle-mounted blood pressure monitor in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means as the method for realizing intelligent monitoring of patient status based on vehicle-mounted sphygmomanometer described in the text can produce the same technical effects, so they will not be repeated here.

[0085] like Figure 3 , which is a schematic diagram of the structure of an electronic device for implementing an intelligent monitoring method of a patient's status based on a vehicle-mounted sphygmomanometer provided by an embodiment of the present invention.

[0086] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a program for an intelligent monitoring method for a patient's status based on a vehicle-mounted blood pressure monitor.

[0087] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the program of the intelligent monitoring method for realizing the patient's status based on the vehicle-mounted blood pressure monitor, but also can be used to temporarily store data that has been output or is to be output.

[0088] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (for example, a program for intelligent monitoring of patient status based on a vehicle-mounted blood pressure monitor, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0089] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0090] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0091] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0092] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0093] Optionally, the electronic device 1 may further include a patient interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visualized patient interface.

[0094] The program of the intelligent monitoring method for patient status based on the vehicle-mounted blood pressure monitor stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve: Using a pre-built pressure sensor, a steering wheel pressure matrix is ​​obtained, and according to the steering wheel pressure matrix, the palm position of the patient is identified to obtain palm position information; Using a pre-built infrared sensor, according to the palm position information, an infrared irradiation operation based on a preset initial infrared wave is performed on the palm position, and a reflection signal of the infrared irradiation operation is received to obtain a palm infrared reflection wave; Using a pre-built vibration sensor, a vibration wave on a pre-built steering wheel is obtained, filtering the palm infrared reflection wave according to the vibration wave to obtain a de-noised palm reflection wave, and performing a difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain an absorbed wave; Using a pre-trained variable pressure blood pressure recognition model, performing time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, performing a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and performing a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution; Performing a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; Performing a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and performing a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; Obtain a health threshold, and determine whether the blood pressure recognition result is greater than the health threshold; When the blood pressure recognition result is greater than the health threshold, the patient is determined to be driving dangerously and an alarm prompt message is generated.

[0095] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0096] Furthermore, if the module / unit integrated in the electronic device 1 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. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0097] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement: Using a pre-built pressure sensor, a steering wheel pressure matrix is ​​obtained, and according to the steering wheel pressure matrix, the palm position of the patient is identified to obtain palm position information; Using a pre-built infrared sensor, according to the palm position information, an infrared irradiation operation based on a preset initial infrared wave is performed on the palm position, and a reflection signal of the infrared irradiation operation is received to obtain a palm infrared reflection wave; Using a pre-built vibration sensor, a vibration wave on a pre-built steering wheel is obtained, filtering the palm infrared reflection wave according to the vibration wave to obtain a de-noised palm reflection wave, and performing a difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain an absorbed wave; Using a pre-trained variable pressure blood pressure recognition model, performing time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, performing a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and performing a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution; Performing a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; Performing a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and performing a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; Obtain a health threshold, and determine whether the blood pressure recognition result is greater than the health threshold; When the blood pressure recognition result is greater than the health threshold, the patient is determined to be driving dangerously and an alarm prompt message is generated.

[0098] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0099] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0100] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0101] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An intelligent monitoring method for patient status based on a vehicle-mounted sphygmomanometer, characterized in that: The method comprises: Using a pre-built pressure sensor, a steering wheel pressure matrix is ​​obtained, and according to the steering wheel pressure matrix, the palm position of the patient is identified to obtain palm position information; Using a pre-built infrared sensor, according to the palm position information, an infrared irradiation operation based on a preset initial infrared wave is performed on the palm position, and a reflection signal of the infrared irradiation operation is received to obtain a palm infrared reflection wave; Using a pre-built vibration sensor, a vibration wave on a pre-built steering wheel is obtained, filtering the palm infrared reflection wave according to the vibration wave to obtain a de-noised palm reflection wave, and performing a difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain an absorbed wave; Using a pre-trained variable pressure blood pressure recognition model, performing time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, performing a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and performing a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution; Performing a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; Performing a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and performing a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; Obtaining a health threshold, and determining whether the blood pressure recognition result is greater than the health threshold; When the blood pressure recognition result is greater than the health threshold, the patient is determined to be driving dangerously and an alarm prompt message is generated.

2. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 1, characterized in that: Before obtaining the steering wheel pressure matrix by using the pre-built pressure sensor, the method further includes: Get the patient's real-time location using pre-built in-car navigation; Obtaining the patient's historical common route, and determining whether the patient is in a straight driving section according to the historical common route and the real-time location; When the patient is in a straight driving section, it is determined that the patient is in a stable driving stage; When the patient is not in a straight driving section, it is determined that the patient is in an unstable driving stage; If the patient is in a stable driving stage, the operation of using the pre-built pressure sensor to obtain the steering wheel pressure matrix is ​​performed.

3. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 2, characterized in that: The step of identifying the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information includes: Performing a masking operation of a preset pressure on the steering wheel pressure matrix to obtain a pressure masking result; Performing an outermost contour edge recognition operation on the pressure masking result to obtain a hand shape distribution of the user; Performing a key node recognition operation on the hand shape distribution to obtain a key node set; The palm position is predicted according to the relative position of each key node in the key node set to obtain the palm position information.

4. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 3, characterized in that: The filtering operation is performed on the palm infrared reflected wave according to the vibration wave to obtain the de-noised palm reflected wave, including: Using a pre-built high-pass filter to perform DC component filtering on the palm infrared reflected wave to obtain a DC component-free wave; The DC component-removed wave is subjected to standardization processing based on a mean of zero and a variance of one to obtain an infrared standard wave; Using a pre-built fast Fourier transform algorithm, the vibration component of the vibration wave is identified to obtain a main vibration frequency; The infrared standard wave is filtered according to the main vibration frequency to obtain a de-noised palm reflection wave.

5. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 4, characterized in that: The variable pressure blood pressure identification model is expressed as: In the formula, represents the blood pressure recognition result, represents the primary blood pressure recognition result, represents the initial weight of the primary blood pressure recognition result for the blood pressure recognition result, represents the steering wheel pressure matrix, Represents the steering wheel pressure matrix Zhong The weight coefficient at the point position, Represents the steering wheel pressure matrix The horizontal axis in Represents the steering wheel pressure matrix The vertical coordinate in Indicates the bias of blood pressure recognition results, represents the bias of the primary blood pressure recognition result, Represents the waveform characteristics, represents the frequency characteristics, Represents the amplitude characteristics, , and All of these are characteristics of the red blood cell flow distribution. , and Represents the weight coefficient.

6. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 5, characterized in that: Before using the pre-trained variable pressure blood pressure recognition model, the method further includes: Obtaining a pre-built variable pressure blood pressure recognition model and obtaining a real blood pressure measurement value sample, wherein the real blood pressure measurement value sample includes a measured hand pressure distribution, a hand blood pressure value label and a real blood pressure value label; Sequentially acquiring a target sample from the real blood pressure measurement value samples, performing hand blood pressure prediction on the target sample using the variable pressure blood pressure recognition model to obtain a hand blood pressure prediction result, and performing real blood pressure prediction on the hand blood pressure result according to the measured hand pressure distribution in the target sample to obtain a real blood pressure prediction result; Using a pre-built cross entropy loss algorithm, the loss value between the hand blood pressure value label of the target sample and the hand blood pressure prediction result is calculated to obtain a first loss value, and the loss value between the true blood pressure value label of the target sample and the true blood pressure prediction result is calculated to obtain a second loss value; Performing weighted calculation on the first loss value and the second loss value to obtain a comprehensive loss value; According to the gradient descent algorithm, the comprehensive loss value is minimized to obtain the network model parameters when the comprehensive loss value is the minimum, and the network model parameters are subjected to a reverse parameter update operation to obtain an updated variable pressure blood pressure recognition model; Return to the above step of obtaining a target sample from the real blood pressure measurement value samples in sequence, and iteratively train the updated variable-pressure blood pressure recognition model. When the sample traversal in the updated variable-pressure blood pressure recognition model is completed, a trained variable-pressure blood pressure recognition model is obtained.

7. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 6, characterized in that: The step of performing time-frequency analysis on the absorbed wave to obtain a set of absorbed wave characteristics includes: Performing a waveform feature extraction operation on the absorbed wave to obtain a time domain signal feature set; Using a pre-built short-time Fourier transform algorithm, the absorbed wave is subjected to a frequency domain conversion operation to obtain a frequency domain signal, and waveform feature extraction is performed on the frequency domain signal to obtain a frequency domain signal feature set; The time domain signal feature set and the frequency domain signal feature set are output to obtain an absorbed wave feature set.

8. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 7, characterized in that: After obtaining the blood pressure recognition result, the method further includes: Recording the blood pressure recognition result to obtain a blood pressure change curve; Performing differential calculation on the blood pressure change curve to obtain a blood pressure change rate curve, and monitoring the blood pressure change rate curve according to a preset warning threshold; When the blood pressure change in the blood pressure change rate curve is greater than the warning threshold, the patient is determined to be driving dangerously, and the vehicle is automatically braked according to a preset emergency braking strategy.

9. The method for intelligently monitoring the patient's status based on a vehicle-mounted sphygmomanometer according to claim 8, characterized in that: The obtaining of the health threshold comprises: Use the pre-built vehicle-mounted central control device to obtain the patient's identity information; Querying a specific information set of the patient according to the identity information, wherein the specific information set includes age, complications and treatment history; According to the specific information set, a preset default hypertension threshold is weightedly calculated to obtain a healthy threshold.

10. An intelligent monitoring system for patient status based on a vehicle-mounted sphygmomanometer, characterized in that: The system comprises: The infrared monitoring module is used to obtain a steering wheel pressure matrix using a pre-built pressure sensor, and identify the palm position of the patient according to the steering wheel pressure matrix to obtain palm position information, and use the pre-built infrared sensor to perform an infrared irradiation operation based on a preset initial infrared wave on the palm position according to the palm position information, and receive a reflection signal of the infrared irradiation operation to obtain a palm infrared reflection wave; The blood absorption wave identification module is used to obtain the vibration wave on the pre-constructed steering wheel by using the pre-constructed vibration sensor, filter the palm infrared reflection wave according to the vibration wave to obtain the de-noised palm reflection wave, and perform difference calculation between the de-noised palm reflection wave and the initial infrared wave to obtain the absorbed wave; A palm blood pressure recognition module is used to use a pre-trained variable pressure blood pressure recognition model to perform time-frequency analysis on the absorbed wave to obtain an absorbed wave feature set, and to perform a classification operation on the absorbed feature set based on a preset red blood cell peak feature to obtain a red blood cell absorption wave feature set, and to perform a red blood cell flow velocity and width distribution recognition operation on the red blood cell absorption wave feature set to obtain a red blood cell flow distribution, and to perform a hand blood pressure recognition operation on the red blood cell flow distribution to obtain a primary blood pressure recognition result; a blood pressure correction module, configured to perform a feature extraction operation on the steering wheel pressure matrix to obtain a palm pressure distribution feature set, and perform a pressure correction operation on the primary blood pressure recognition result according to the palm pressure distribution feature set to obtain a blood pressure recognition result; The supervision and warning module is used to obtain a health threshold, determine whether the blood pressure recognition result is greater than the health threshold, and when the blood pressure recognition result is greater than the health threshold, determine that the patient is driving dangerously and generate an alarm prompt message.

Images