Blood pressure measuring device of an electronic sphygmomanometer

Abstract

An electronic blood pressure monitor includes a measurement unit and a sensor module. The sensor module comprises a transmitting module and a receiving module, respectively mounted on the user's two wrists. The measurement unit includes a processor, a pressure sensor, and a display screen. The processor receives and processes signals transmitted from the sensor module and calculates the user's blood pressure value using an improved neural network. The pressure sensor measures the pressure signal at the user's wrist, and the display screen shows the measurement results. This invention eliminates the need for an armband: the sensor module is directly mounted on the user's wrist, avoiding errors caused by improper fixation and significantly improving the accuracy of blood pressure measurement, thereby enhancing the user experience.

Description

Technical Field

[0001] This invention relates to the field of blood pressure measurement technology, and more specifically to a blood pressure measuring device for an electronic blood pressure monitor. Background Technology

[0002] Currently, electronic blood pressure monitors, as non-invasive blood pressure measurement devices, are widely used in clinical and home health monitoring. Traditional electronic blood pressure monitors typically consist of an arm cuff and a measuring unit. The arm cuff measures the user's blood pressure by inflating and deflating the cuff. However, existing electronic blood pressure monitors have some problems during measurement, such as the need for proper arm cuff fixation, which can lead to errors if used improperly; relatively slow measurement speed, making them unsuitable for high-frequency continuous measurements; and the arm cuff design and usage are sometimes not comfortable or convenient enough.

[0003] Furthermore, existing blood pressure monitors are inaccurate due to various factors, and there are no new devices or methods that adopt artificial intelligence-related improvements for measurement, making them unable to adapt to the current situation where accurate and rapid blood pressure measurement is required. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a blood pressure measuring device for an electronic blood pressure monitor. This device significantly improves measurement accuracy and convenience, greatly enhancing accuracy and improving the user experience.

[0005] An electronic blood pressure monitor includes a measurement unit and a sensor module. The sensor module includes a transmitting module and a receiving module, which are respectively installed on the user's two wrists. The measurement unit includes a processor, a pressure sensor, and a display screen. The processor receives and processes signals transmitted from the sensor module and calculates the user's blood pressure value using an improved neural network. The pressure sensor measures the pressure signal at the user's wrist. The display screen displays the measurement results. After acquiring the pulse wave signal, the pressure sensor performs preprocessing to remove noise. Simultaneously, it detects the peak value, trough value, rising edge, and falling edge slope characteristics of the pulse wave signal to determine individual characteristics: age, gender, and body mass index (BMI). Based on the characteristics of the pulse wave signal and the user's individual characteristics, the improved neural network is used to predict and calculate the blood pressure value, outputting T. T is corrected based on the user's individual characteristics to obtain a blood pressure correction value. The corrected blood pressure value is then sent to the display screen for display.

[0006] Preferably, the sensor module employs wireless transmission technology, transmitting data via high-frequency signals.

[0007] Preferably, the sensor module includes a pulse sensor capable of measuring pulse wave signals at the user's wrist.

[0008] Preferably, the measuring unit has an automatic start and stop function. The measurement is automatically started when the user places his wrist on the sensor module and automatically stops when the measurement is completed.

[0009] Preferably, the measuring unit has a storage function, which can record and save multiple measurement results, making it convenient for users to monitor and manage their health status.

[0010] Preferably, the improved neural network uses historical pulse wave signal data and corresponding known blood pressure values ​​as training datasets, dividing the dataset into input features and target outputs. The input features specifically include peak values, trough values, the slope of the rising and falling edges, age, gender, and body mass index. The target outputs specifically include systolic and diastolic blood pressure. The improved hidden layer sigmoid function used in the improved neural network is:

[0011]

[0012] The improved output layer Sigmoid function used in the improved neural network is:

[0013]

[0014] Where e is the natural index, x is the input, and K is the blood pressure correction coefficient.

[0015] Preferably, the sensor module and the measuring unit transmit data wirelessly, avoiding the inconvenience of connecting wires required in traditional arm-cuff blood pressure monitors.

[0016] Preferably, the wireless transmission between the sensor module and the measurement unit employs encryption technology to ensure secure data transmission.

[0017] Preferably, the wireless transmission between the sensor module and the measuring unit employs low-power technology, which extends the lifespan of the electronic blood pressure monitor.

[0018] Preferably, the step of correcting the blood pressure value output by the improved neural network by incorporating the user's individual characteristics includes:

[0019] K = m * age + n * gender + r * BMI + intercept

[0020] Where K is the blood pressure correction coefficient, m is the age regression coefficient, n is the gender regression coefficient, and r is the BMI regression coefficient. These are set based on historical experience. Individual characteristics can be obtained through user input, body sensors, and health records. The obtained blood pressure correction coefficient is multiplied by T to obtain the blood pressure correction value.

[0021] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0022] (1) No armband required: The sensor module is directly mounted on the user's wrist, eliminating the need for an armband and avoiding errors caused by improper mounting. Suitable for continuous measurement: Users only need to place both wrists on the sensor module to achieve continuous blood pressure measurement, suitable for occasions requiring high-frequency measurements; Convenient and comfortable to use: Users only need to place their wrists on the sensor module, without complicated armband adjustments, making operation simple and comfortable; Fast measurement speed: The sensor module transmits wireless signals, resulting in a faster measurement speed than traditional armband-type electronic blood pressure monitors, improving measurement efficiency. Suitable for various users: Traditional armband-type electronic blood pressure monitors may not be suitable for certain groups, such as obese individuals, children, and the elderly. The blood pressure measuring device of this invention can be flexibly mounted on the wrist, suitable for users of various body types and ages.

[0023] (2) Accurate and reliable: The blood pressure at the user's wrist is measured by the sensor module, avoiding the errors that may occur in traditional blood pressure monitors with arm cuffs, and providing more accurate and reliable measurement results; the blood pressure value output by the improved neural network is corrected by: blood pressure correction coefficient K = m * age + n * gender + r * BMI + intercept, where m is the age regression coefficient, n is the gender regression coefficient, and r is the BMI regression coefficient, which is set according to historical experience. Individual characteristics can be obtained through user input, body sensors, and health records. The obtained blood pressure correction coefficient is multiplied by T to obtain the blood pressure correction value.

[0024] (3) Intuitive display: The measuring unit is equipped with a display screen, which can intuitively display the user's blood pressure value, making it convenient for the user to refer to and record.

[0025] (4) An improved neural network is used, wherein the improved hidden layer Sigmoid function of the improved neural network is:

[0026]

[0027] The improved output layer Sigmoid function used in the improved neural network is:

[0028]

[0029] Where e is the natural index, x is the input, and K is the blood pressure correction coefficient. By introducing the blood pressure correction coefficient into the Sigmoid function, individual characteristics are taken into account, which greatly enhances the accuracy of blood pressure measurement and improves the user experience. Attached Figure Description

[0030] Figure 1 This is a system diagram of a blood pressure measuring device for an electronic blood pressure monitor according to the present invention; Detailed Implementation

[0031] Those skilled in the art will understand that, as mentioned in the background section, traditional blood pressure measurements have low accuracy and cannot be combined with historical measurement data, making them incompatible with existing voltage measurements. Existing technologies only involve voltage measurement and do not consider artificial intelligence technology. Achieving a high degree of integration and matching between voltage measurement and real-world needs is crucial for successful voltage measurement. How to make voltage measurement more intelligent and user-friendly, improve its operational efficiency and accuracy, and enhance user readability has become a new research topic. To make the above-mentioned objectives, features, and beneficial effects of this invention more apparent and understandable, specific embodiments of the invention will be described in detail below with reference to the accompanying drawings.

[0032] Example 1:

[0033] Figure 1 This illustration shows a system diagram of a blood pressure measuring device for an electronic blood pressure monitor according to this application. The device includes a measuring unit and a sensor module. In one embodiment, the measuring unit may employ a microprocessor chip as the processor to receive and process signals transmitted from the sensor module and calculate the user's blood pressure value using an algorithm. A high-precision pressure sensor may be selected to measure the pressure signal at the user's wrist. The display screen may be an LCD screen or other suitable display screen to display the user's blood pressure value. The sensor module includes a transmitting module and a receiving module, respectively installed on the user's two wrists. The measuring unit includes a processor, a pressure sensor, and a display screen. The processor receives and processes signals transmitted from the sensor module and calculates the user's blood pressure value using an improved neural network. The pressure sensor measures the pressure signal at the user's wrist, and the display screen displays the measurement results.

[0034] It also includes a transmitting module and a receiving module, which are installed on the user's two wrists respectively. The transmitting module sends high-frequency signals to the receiving module via wireless transmission technology to measure the blood pressure at the user's wrist. After receiving the signal, the receiving module uses built-in sensors to measure the pulse wave signal and pressure signal at the user's wrist, and transmits these signals back to the measurement unit for processing.

[0035] In use, the user simply places their wrist on the sensor module to activate the measurement unit. The unit then continuously measures blood pressure at the user's wrist via the sensor module. Upon receiving the signal from the sensor module, the unit's processor calculates the user's blood pressure using a built-in algorithm and displays it on the screen. Users can record or save the measurement results as needed.

[0036] In practical applications, the blood pressure measuring device of the electronic blood pressure monitor of this invention can be widely used in clinical medicine, home health monitoring, and other fields. For example, in medical institutions such as hospitals and clinics, doctors or nurses can use this device to perform rapid and continuous blood pressure measurements on patients and monitor the trend of blood pressure changes. At home, users can conveniently measure their own blood pressure using this device, understand their health status, and report the measurement results to the doctor in a timely manner so that the treatment plan can be adjusted accordingly.

[0037] Furthermore, after the pressure sensor acquires the pulse wave signal, it performs preprocessing to remove noise; at the same time, it detects the peak value, trough value, rising edge and falling edge slope characteristics of the pulse wave signal to determine individual characteristics: age, gender, and body mass index (BMI).

[0038] This technical solution uses an improved neural network to predict and calculate blood pressure value T based on the characteristics of pulse wave signals and the individual characteristics of users. The T value is then corrected to obtain a blood pressure correction value, which is then sent to the display screen for display.

[0039] In some embodiments, the sensor module employs wireless transmission technology to transmit data via high-frequency signals. This offers advantages such as portability, wireless connectivity, automatic start and stop, high-precision measurement, and data transmission and storage. A pulse sensor at the wrist measures the pulse wave signal, and a high-precision algorithm from a processor calculates the user's blood pressure. The measurement result is then wirelessly transmitted to the display screen of the measurement unit for display to the user. Multiple measurement results can also be saved for health monitoring and management.

[0040] Furthermore, the blood pressure measuring device of the electronic blood pressure monitor of this invention features wireless transmission encryption technology to ensure data transmission security. It also employs low-power technology to extend the lifespan of the electronic blood pressure monitor and enhance the user experience. Moreover, compared to traditional arm-cuff blood pressure monitors, this invention avoids the inconvenience of connecting cables, making it more portable and suitable for daily life and mobile use.

[0041] Therefore, the electronic blood pressure monitor of the present invention has significant advantages in terms of convenience, accuracy, data transmission, and storage, and has broad application prospects. The electronic blood pressure monitor of the present invention is suitable for multiple fields, including medical, health management, and personal / home use, and has broad market application potential.

[0042] In some embodiments, the sensor module includes a pulse sensor capable of measuring pulse wave signals at the user's wrist.

[0043] In some embodiments, the measuring unit has an automatic start and stop function. The measurement is automatically started when the user places his wrist on the sensor module and automatically stops when the measurement is completed.

[0044] In this technical solution, pulse wave signals are acquired and preprocessed, such as by removing noise and filtering, to ensure the accuracy and reliability of the signals.

[0045] In this technical solution, key features of the pulse wave signal are detected, such as the peak value, trough value, and slope of the rising and falling edges of the pulse wave.

[0046] Based on the characteristics of pulse wave signals, complex mathematical models, such as neural networks and support vector machines, are used to predict and calculate blood pressure values. These mathematical models can be trained and optimized using a large amount of actual measurement data, thereby improving the accuracy and precision of the algorithm.

[0047] Further calibration and correction are performed based on the predicted and calculated blood pressure values. For example, individual user characteristics such as age, gender, and body mass index (BMI) can be incorporated for personalized correction to improve the accuracy and precision of the measurement results.

[0048] The calculated systolic and diastolic blood pressure values ​​are displayed on the measurement unit's screen for user viewing.

[0049] Here is a simplified mathematical representation of a possible high-precision algorithm example:

[0050] SBP = f(peak value, trough value, slope, age, gender, BMI)

[0051] DBP = g(peak value, trough value, slope, age, gender, BMI)

[0052] Here, f and g represent functions for predicting and calculating systolic and diastolic blood pressure, respectively; peak value, trough value, and slope are characteristics of the pulse wave signal; and age, gender, and BMI are individual user characteristics. These functions can be adjusted and optimized based on actual measurement data and model training to improve the accuracy and precision of the algorithm.

[0053] It should be noted that the above algorithm is only an example. The actual high-precision algorithm can be further optimized and improved according to the actual situation and needs to ensure the accuracy and reliability of the measurement results.

[0054] In some embodiments, the measuring unit has a storage function, which can record and save multiple measurement results, facilitating users to monitor and manage their health status.

[0055] In some embodiments, the improved neural network uses historical pulse wave signal data and corresponding known blood pressure values ​​as training datasets, dividing the datasets into input features and target outputs. The input features specifically include peak values, trough values, the slope of the rising and falling edges, age, gender, and body mass index. The target outputs specifically include systolic and diastolic blood pressure. The improved hidden layer sigmoid function used in the improved neural network is:

[0056]

[0057] The improved output layer Sigmoid function used in the improved neural network is:

[0058]

[0059] Where e is the natural exponent, x is the input, and K is the blood pressure correction coefficient. Data input: The collected pulse wave signal features are used as the input to the neural network, and then fed into the network's input layer.

[0060] Network forward propagation: The forward propagation of the network includes weighting and activation function processing of the input layer to obtain the output of the hidden layer.

[0061] Hidden layer processing: The Sigmoid activation function can be used in the hidden layer to restrict the output of the hidden layer to between 0 and 1.

[0062] Network output layer processing: The output of the hidden layer is taken as input, and processed by the weighted sum and activation function of the output layer to obtain the final blood pressure value prediction.

[0063] Correction and adjustment: Based on the user's individual characteristics, such as age, gender, BMI, etc., the predicted blood pressure value can be corrected and adjusted through linear regression or other mathematical models to obtain the final blood pressure value.

[0064] In some embodiments, the sensor module and the measuring unit transmit data wirelessly, avoiding the inconvenience of connecting wires required in traditional arm-cuff blood pressure monitors.

[0065] In some embodiments, the wireless transmission between the sensor module and the measurement unit employs encryption technology to ensure secure data transmission. The following is a process for predicting and calculating blood pressure values ​​using a neural network:

[0066] Data preparation: Collect a large amount of pulse wave signal data and corresponding known blood pressure values ​​as training datasets. Divide the dataset into input features (such as peak value, trough value, slope, etc.) and target outputs (systolic blood pressure and diastolic blood pressure).

[0067] Neural network architecture design: Choose a suitable neural network architecture, such as a multi-layer perceptron (MLP) or a convolutional neural network (CNN), and define the number of neurons and activation functions of the input layer, hidden layer, and output layer.

[0068] Model training: The neural network is trained using a training dataset. The network weights and biases are continuously adjusted through the backpropagation algorithm, enabling the network to learn the correlation between pulse wave signal characteristics and blood pressure values.

[0069] Model validation: The trained model is validated using a validation dataset to evaluate its performance, such as accuracy, recall, and F1 score, to ensure the model's predictive and computational capabilities.

[0070] Model prediction and calculation: The trained and validated neural network can be used to predict and calculate a user's blood pressure. The features of the input pulse wave signal are used as input to the network, and after forward propagation, the model outputs the predicted systolic and diastolic blood pressure values.

[0071] The following is a simplified formula, assuming the neural network has one input layer, one hidden layer, and one output layer, using the sigmoid activation function:

[0072] Input features: X = [peak value, trough value, slope, ...] (vector form)

[0073] Hidden layer weights: W1

[0074] Hidden layer bias: B1

[0075] Hidden layer output: A1 = sigmoid(W1*X + B1)

[0076] Output layer weights: W2

[0077] Output layer bias: B2

[0078] Output layer output: A2 = sigmoid(W2*A1+B2)

[0079] Predicted systolic blood pressure: SBP = A2[0]

[0080] Predicted diastolic blood pressure: DBP = A2[1]

[0081] Wherein, sigmoid represents the sigmoid activation function, W1, B1, W2, and B2 represent the weights and biases of the hidden and output layers, respectively, and A1 and A2 represent the outputs of the hidden and output layers, respectively. By adjusting the weights and biases of the neural network, A2[0] and A2[1] can be made to represent the predicted systolic and diastolic blood pressure values, respectively. Other high-precision algorithms, such as Support Vector Machine (SVM) and deep learning models (such as recurrent neural networks and long short-term memory networks), can also be combined to select appropriate algorithms for blood pressure prediction and calculation based on the actual situation. These algorithms have their own advantages and applicable scenarios, and can be selected and adjusted according to specific needs.

[0082] In some embodiments, the wireless transmission between the sensor module and the measurement unit employs low-power technology, extending the lifespan of the electronic blood pressure monitor.

[0083] In some embodiments, correcting the blood pressure value output by the improved neural network incorporating the user's individual characteristics includes:

[0084] K = m * age + n * gender + r * BMI + intercept

[0085] Where K is the blood pressure correction coefficient, m is the age regression coefficient, n is the gender regression coefficient, r is the BMI regression coefficient, and the intercept is the intercept term of the model, which is set according to historical experience. Individual characteristics can be obtained through user input, body sensors, and health records. The obtained blood pressure correction coefficient is multiplied by T to obtain the blood pressure correction value.

[0086] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0087] (1) No armband required: The sensor module is directly mounted on the user's wrist, eliminating the need for an armband and avoiding errors caused by improper mounting. Suitable for continuous measurement: Users only need to place both wrists on the sensor module to achieve continuous blood pressure measurement, suitable for occasions requiring high-frequency measurements; Convenient and comfortable to use: Users only need to place their wrists on the sensor module, without complicated armband adjustments, making operation simple and comfortable; Fast measurement speed: The sensor module transmits wireless signals, resulting in a faster measurement speed than traditional armband-type electronic blood pressure monitors, improving measurement efficiency. Suitable for various users: Traditional armband-type electronic blood pressure monitors may not be suitable for certain groups, such as obese individuals, children, and the elderly. The blood pressure measuring device of this invention can be flexibly mounted on the wrist, suitable for users of various body types and ages.

[0088] (2) Accurate and reliable: The blood pressure at the user's wrist is measured by the sensor module, avoiding the errors that may occur in traditional blood pressure monitors with arm cuffs, and providing more accurate and reliable measurement results; the blood pressure value output by the improved neural network is corrected by: blood pressure correction coefficient K = m * age + n * gender + r * BMI + intercept, where m is the age regression coefficient, n is the gender regression coefficient, and r is the BMI regression coefficient, which is set according to historical experience. Individual characteristics can be obtained through user input, body sensors, and health records. The obtained blood pressure correction coefficient is multiplied by T to obtain the blood pressure correction value.

[0089] (3) Intuitive display: The measuring unit is equipped with a display screen, which can intuitively display the user's blood pressure value, making it convenient for the user to refer to and record.

[0090] (4) An improved neural network is used, wherein the improved hidden layer Sigmoid function of the improved neural network is:

[0091]

[0092] The improved output layer Sigmoid function used in the improved neural network is:

[0093]

[0094] Where e is the natural index, x is the input, and K is the blood pressure correction coefficient. By introducing the blood pressure correction coefficient into the Sigmoid function, individual characteristics are taken into account, which greatly enhances the accuracy of blood pressure measurement and improves the user experience.

[0095] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products, and therefore this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0096] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A blood pressure measuring device for an electronic blood pressure monitor, comprising a measuring unit and a sensor module, characterized in that: The sensor module includes a transmitting module and a receiving module, which are respectively installed on the user's two wrists. The measuring unit includes a processor, a pressure sensor, and a display screen. The processor is used to receive and process the signals transmitted back by the sensor module and calculate the user's blood pressure value through an improved neural network. The pressure sensor is used to measure the pressure signal at the user's wrists, and the display screen is used to display the measurement results. After the pressure sensor acquires the pulse wave signal, it undergoes preprocessing to remove noise. Simultaneously, the peak value, trough value, rising edge, and falling edge slope characteristics of the pulse wave signal are detected to determine individual characteristics: age, gender, and body mass index (BMI). Specifically, individual characteristics are obtained through user input, body sensors, and health records. Based on the characteristics of pulse wave signals and individual user characteristics, an improved neural network is used to predict blood pressure values ​​and calculate the output T. The improved neural network is a multilayer perceptron, and the number of neurons and activation functions of the input layer, hidden layer and output layer are defined. Forward propagation of the network: The forward propagation of the network includes weighting and activation function processing of the input layer to obtain the output of the hidden layer; Hidden layer processing: Use the Sigmoid activation function in the hidden layer to restrict the output of the hidden layer to between 0 and 1; Network output layer processing: The output of the hidden layer is taken as input, and processed by the weighted summation and activation function of the output layer to obtain the final blood pressure value prediction; The features of the input pulse wave signal are used as the input to the network. After forward propagation, the output of the model is obtained, which is the predicted systolic and diastolic blood pressure values. The network weights and biases are continuously adjusted through the backpropagation algorithm, so that the network can learn the correlation between the pulse wave signal features and blood pressure values. The blood pressure correction value is obtained by adjusting T based on the user's individual characteristics; specifically, the blood pressure correction coefficient is multiplied by T to obtain the blood pressure correction value. Send the corrected blood pressure value to the display screen; The improved neural network uses historical pulse wave signal data and corresponding known blood pressure values ​​as training datasets. The dataset is divided into input features and target outputs. Input features specifically include peak values, trough values, slope features of the rising and falling edges, age, gender, and body mass index. The target outputs specifically include systolic and diastolic blood pressure. The improved hidden layer sigmoid function used in the improved neural network is as follows: The improved output layer Sigmoid function used in the improved neural network is: Where e is the natural index, x is the input, and K is the blood pressure correction coefficient; The step of correcting the blood pressure value output by the improved neural network by incorporating the user's individual characteristics includes: K = m * age + n * gender + r * BMI + intercept Where K is the blood pressure correction coefficient, m is the age regression coefficient, n is the gender regression coefficient, and r is the BMI regression coefficient, which are set based on historical experience.

2. The blood pressure measuring device of the electronic blood pressure monitor according to claim 1, characterized in that, The sensor module uses wireless transmission technology to transmit data via high-frequency signals.

3. The blood pressure measuring device of the electronic blood pressure monitor according to claim 1, characterized in that, The measurement unit has an automatic start and stop function. The measurement will start automatically when the user places his wrist on the sensor module and will stop automatically after the measurement is completed.

4. The blood pressure measuring device of the electronic blood pressure monitor according to claim 1, characterized in that, The measurement unit has a storage function, which can record and save multiple measurement results, making it convenient for users to monitor and manage their health status.

5. The blood pressure measuring device of the electronic blood pressure monitor according to claim 1, characterized in that, The sensor module and the measurement unit transmit data wirelessly.

6. The blood pressure measuring device of the electronic blood pressure monitor according to claim 5, characterized in that, The wireless transmission between the sensor module and the measurement unit employs encryption technology.

7. The blood pressure measuring device of the electronic blood pressure monitor according to claim 5, characterized in that, The wireless transmission between the sensor module and the measurement unit employs low-power technology.

Images