Self-adaptive pneumothorax pressure and flow intelligent control monitoring system and method

Abstract

The invention relates to a self-adaptive pneumothorax pressure and flow intelligent control monitoring system and method. The system comprises a sensor module, a central processing module and an execution module. The central processing module is in communication connection with the sensor module and the execution module; the sensor module comprises a pressure sensor and a flow sensor, and the pressure sensor is used for monitoring air pressure change in the thoracic cavity in real time; the flow sensor is used for measuring the flow of gas passing through the drainage tube; the execution module is a drainage valve, adjusts the opening degree of the drainage valve according to the control signal of the central processing module, and controls the gas exhaust speed; a fuzzy logic control algorithm and a neural network training model are arranged in the central processing module; the fuzzy logic control algorithm and the neural network training model are combined to carry out data processing analysis and prediction, real-time monitoring is carried out, and the opening degree of the drainage valve is controlled. Compared with the prior art, pressure and flow changes in the thoracic cavity of a patient can be monitored and responded in real time, and accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to an adaptive pneumothorax pressure and flow intelligent control monitoring system and method. Background Art

[0002] Pneumothorax is a common and serious respiratory disease that may be caused by trauma, disease or surgery. It causes partial or complete collapse of the lungs, affecting normal respiratory function and can be life-threatening in severe cases. The clinical manifestations of pneumothorax include sudden chest pain, dyspnea, shortness of breath and other symptoms, which require immediate medical intervention.

[0003] Currently, one of the main methods for treating pneumothorax is closed chest drainage, which uses negative pressure to draw air out of the chest cavity by placing a drainage tube in the patient's chest cavity to relieve lung compression and promote lung re-expansion. Although this method is widely used in clinical practice, it has some limitations:

[0004] 1. Insufficient monitoring: Traditional drainage devices usually lack accurate monitoring functions and cannot reflect the dynamic changes of intrathoracic pressure and drainage flow in real time. This makes it difficult for doctors to accurately evaluate the treatment effect and the real-time changes of the patient's condition.

[0005] 2. Manual adjustment: Due to the lack of an automated control system, doctors need to manually adjust the parameters of the drainage device, such as drainage rate and negative pressure level, based on experience and regular inspections. This is not only inefficient but may also lead to inaccurate treatment due to human factors.

[0006] 3. Safety risks: The inability to monitor intrathoracic pressure in real time may put patients at risk of over-drainage or under-drainage, which may cause further complications such as recurrent pneumothorax, infection, or lung damage.

[0007] 4. Treatment efficiency: Due to the lack of automation and intelligence, medical staff need to frequently monitor the patient's condition, which increases the workload of medical staff and reduces treatment efficiency.

[0008] In view of these limitations, it is particularly urgent to develop an adaptive pneumothorax pressure and flow intelligent control monitoring device to help medical staff make timely clinical decisions and improve the accuracy and safety of treatment.

[0009] Patent CN202410545518.7 provides a monitoring device for closed chest drainage. A specific embodiment of the device includes: the monitoring device for closed chest drainage includes: a closed chest drainage device and a monitoring device, the closed chest drainage device includes: a catheter holder, a three-way connector, a three-way rotary valve, a drainage catheter, a transparent drainage bag, and the monitoring device includes: an image acquisition device and an image analysis processor; the image analysis processor is configured to: input the drainage bag frame image sequence into a pre-trained drainage bag cap detection model; input the drainage bag frame image into a pre-trained drainage bag water column detection model; send an alarm prompt message to the associated alarm device. This embodiment can promptly prompt medical staff to check on the patient through an alarm device when abnormal chest drainage is detected. However, it cannot solve the technical problem of real-time monitoring and response to changes in pressure and flow in the patient's chest cavity to improve accuracy. Summary of the invention

[0010] The purpose of the present invention is to overcome the above-mentioned defects of the prior art and provide an adaptive pneumothorax pressure and flow intelligent control monitoring device and method.

[0011] The present invention realizes precise monitoring through intelligent control, and provides personalized diagnosis and treatment schemes for patients with closed chest drainage. The present invention is based on sensor technology, microelectronics technology, control theory and artificial intelligence algorithm to realize real-time monitoring and intelligent control of intrathoracic pressure and drainage flow of patients with pneumothorax. Data is collected by high-precision sensors, the central processing unit uses intelligent algorithms to process data and generate control signals, and the actuator automatically adjusts the drainage system according to the control signal to achieve precise treatment.

[0012] The present invention provides an adaptive pneumothorax pressure and flow intelligent control monitoring system, comprising: a sensor module, a central processing module, and an execution module; the central processing module is communicatively connected with the sensor module and the execution module;

[0013] The sensor module is arranged near the chest cavity of the patient, and includes a pressure sensor and a flow sensor. The pressure sensor is used to monitor the air pressure changes in the chest cavity in real time, and convert the data into an electrical signal and send it to the central processing module; the flow sensor is used to measure the gas flow through the drainage tube; the execution module is a drainage valve, and the execution module automatically adjusts the opening of the drainage valve according to the control signal of the central processing module to control the gas discharge speed;

[0014] The central processing module includes a microprocessor and a storage unit. The microprocessor is used to receive signals from the sensor module and execute a pre-programmed algorithm to analyze data. The storage unit is used to save the patient's treatment history, parameter settings and algorithm update information; the storage unit is a built-in RAM or an external hard disk drive.

[0015] The central processing module is provided with a fuzzy logic control algorithm and a neural network training model;

[0016] The fuzzy logic control algorithm includes:

[0017] A1: Fuzzification process: Fuzzification is performed on the input variables, which are pressure and flow, and a membership function is defined for each input variable to map the input value to the membership degree in the fuzzy set;

[0018] A2: Establish a fuzzy rule base: Establish fuzzy rules to describe the relationship between input variables and output variables; the output variable is the opening of the drainage valve;

[0019] A3: Fuzzy reasoning: Based on the membership degree of input variables and the fuzzy rule base, fuzzy reasoning is performed to obtain the fuzzy set of output variables;

[0020] A4: Defuzzification: convert the fuzzy set of output variables obtained by fuzzy reasoning into specific numerical output, i.e. the opening of the diversion valve;

[0021] The neural network training model is a feedforward neural network model, which includes an input layer, a hidden layer and an output layer; the pressure and flow change data and the patient's age, weight, and severity of the disease are input into the input layer, the output layer outputs the data for the prediction of the drainage valve opening, the hidden layer extracts complex features from the input data, and introduces nonlinearity through an activation function to achieve prediction;

[0022] The fuzzy logic control algorithm and the neural network training model are combined to perform data processing analysis and prediction, conduct real-time monitoring, and control the opening of the drainage valve.

[0023] Furthermore, in the fuzzy logic control algorithm, in A1, the membership function is a triangular function, a trapezoidal function or a Gaussian function.

[0024] Furthermore, in the fuzzy logic control algorithm, in A3, the fuzzy reasoning method is the Mamdani method and the Sugeno method.

[0025] Furthermore, in the fuzzy logic control algorithm, in A4, the defuzzification method is the centroid method and the maximum membership method.

[0026] Furthermore, in the fuzzy logic control algorithm, in A2, the fuzzy rule base is: if the pressure is large and the flow is small, then the opening of the drainage valve is large; if the pressure is medium and the flow is medium, then the opening of the drainage valve is medium; if the pressure is small and the flow is large, then the opening of the drainage valve is small.

[0027] Furthermore, the training process of the neural network training model is:

[0028] B1: Collect and organize data sets for training and testing, including intrathoracic air pressure, flow, patient age, weight, disease severity, and drainage valve opening, preprocess the data sets, and divide the data sets into training set, validation set, and test set;

[0029] B2: Network structure construction: Determine the number of layers and the number of neurons in each layer, including the number of input layers, hidden layers, and output layers and the number of neurons in each layer, and select activation functions such as Sigmoid, Tanh, ReLU, etc. Determine the loss function, such as mean square error (MSE) or cross-entropy loss (Cross-Entropy Loss).

[0030] B3: Initialize the parameters, then forward propagate, pass the input data to the input layer, hidden layer and output layer for calculation, calculate the activation function, and then calculate the difference between the output and the true value through the loss function, and then use the gradient descent method to update the parameters, repeat forward propagation, loss calculation, back propagation and parameter update until the predetermined number of iterations is reached or the loss function converges;

[0031] B4: Model evaluation, use the validation set to evaluate the performance of the model, adjust the hyperparameters to improve the model effect, and use the test set to evaluate the generalization ability of the model.

[0032] Furthermore, it also includes a user interaction interface that can provide an intuitive user interface, allowing medical staff to easily set and adjust treatment parameters while also being able to monitor patients' real-time data.

[0033] Furthermore, the user interaction interface includes:

[0034] The display unit displays key parameters such as intrathoracic pressure and drainage flow in real time.

[0035] A control panel provides a manual control option, allowing medical staff to manually intervene when necessary.

[0036] The alarm system will automatically sound an alarm when it detects abnormal conditions, such as excessive pressure or abnormal flow.

[0037] Furthermore, the flow sensor is a differential pressure flow sensor, an ultrasonic flow sensor or a thermal flow sensor, which can accurately measure the gas flow velocity.

[0038] The present invention also provides an adaptive pneumothorax pressure and flow intelligent control method, comprising the following steps:

[0039] S1: Installation and calibration of sensor module: Install the pressure sensor near the patient's chest to ensure that it can accurately monitor the intrathoracic pressure; install the flow sensor on the drainage tube to ensure that it can accurately measure the gas flow; calibrate the sensor to ensure its measurement accuracy; insert the drainage tube into the drainage tube placement;

[0040] S2: Configuration of the central processing module: configure the microprocessor to ensure that it can receive and process data from the sensor; configure the control algorithm unit, including the trained fuzzy logic control algorithm and neural network training model;

[0041] S3: Implement closed-loop control. The sensor module receives data in real time and inputs it into the control algorithm. The control algorithm calculates the optimal control signal and sends it to the actuator to adjust the pneumothorax pressure. The control signal is adjusted according to real-time feedback to ensure that the pressure is maintained within the target range.

[0042] S4: User interaction interface, which displays intrathoracic pressure and drainage flow parameters, provides manual control options, and allows medical staff to manually intervene when necessary; implements an alarm system that automatically issues an alarm when an abnormal situation is detected.

[0043] Implement fault detection mechanisms so that problems can be detected and actions can be taken in a timely manner. Design emergency shutdown and manual control options to ensure patient safety and stable operation of equipment.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] (1) Real-time: The system can monitor and respond to changes in pressure and flow in the patient's chest cavity in real time.

[0046] (2) Accuracy: High-precision sensors and advanced control algorithms ensure the accuracy of treatment.

[0047] (3) Intelligence: Intelligent control algorithms can automatically adjust treatment parameters and reduce the workload of medical staff.

[0048] (4) Personalization: Personalized treatment plans based on the patient’s condition to improve treatment outcomes.

[0049] (5) Easy to operate: The user interface is intuitive and easy to use, simplifying the operation process.

[0050] (6) Safety: It has emergency shutdown and fault self-detection functions to ensure patient safety.

[0051] (7) Environmental protection and energy saving: Use environmentally friendly materials and energy-saving design to reduce energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1It is a working diagram of the adaptive pneumothorax pressure and flow intelligent control monitoring system;

[0053] Figure 2 Schematic diagram of the installation location of the sensor module and user interaction interface.

[0054] Figure numerals: 1 - sensor module; 2 - drainage tube; 3 - user interaction interface; 4 - drainage tube placement location. DETAILED DESCRIPTION

[0055] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Component models, material names, connection structures, control methods, algorithms and other features not clearly described in this technical solution are all considered to be common technical features disclosed in the prior art.

[0056] Example 1

[0057] This embodiment provides an adaptive pneumothorax pressure and flow intelligent control monitoring system. Figure 1 , 2 As shown, it includes: a sensor module 1, a central processing module, and an execution module; the central processing module is connected to the sensor module 1 and the execution module in communication;

[0058] The sensor module 1 is arranged near the chest cavity of the patient. The sensor module 1 includes a pressure sensor and a flow sensor. The pressure sensor is used to monitor the air pressure changes in the chest cavity in real time and convert the data into an electrical signal to send to the central processing module; the flow sensor is used to measure the gas flow through the drainage tube 2; the execution module is a drainage valve. The execution module automatically adjusts the opening of the drainage valve according to the control signal of the central processing module to control the gas discharge speed;

[0059] The central processing module includes a microprocessor and a storage unit. The microprocessor is used to receive signals from the sensor module 1 and execute a pre-programmed algorithm to analyze data. The storage unit is used to save the patient's treatment history, parameter settings and algorithm update information; the storage unit is a built-in RAM or an external hard disk drive.

[0060] The central processing module is provided with a fuzzy logic control algorithm and a neural network training model;

[0061] The fuzzy logic control algorithm includes:

[0062] A1: Fuzzification process: Fuzzification is performed on the input variables, which are pressure and flow, and a membership function is defined for each input variable to map the input value to the membership degree in the fuzzy set;

[0063] A2: Establish a fuzzy rule base: Establish fuzzy rules to describe the relationship between input variables and output variables; the output variable is the opening of the drainage valve;

[0064] A3: Fuzzy reasoning: Based on the membership degree of input variables and the fuzzy rule base, fuzzy reasoning is performed to obtain the fuzzy set of output variables;

[0065] A4: Defuzzification: convert the fuzzy set of output variables obtained by fuzzy reasoning into specific numerical output, i.e. the opening of the diversion valve;

[0066] The neural network training model is a feedforward neural network model, which includes an input layer, a hidden layer and an output layer; the pressure and flow change data and the patient's age, weight, and severity of the disease are input into the input layer, the output layer outputs the data for the prediction of the drainage valve opening, the hidden layer extracts complex features from the input data, and introduces nonlinearity through an activation function to achieve prediction;

[0067] The fuzzy logic control algorithm and the neural network training model are combined to perform data processing analysis and prediction, conduct real-time monitoring, and control the opening of the drainage valve.

[0068] In a specific implementation, in the fuzzy logic control algorithm, in A1, the membership function is a triangular function, a trapezoidal function or a Gaussian function.

[0069] In a specific implementation, in the fuzzy logic control algorithm, in A3, the fuzzy reasoning method is the Mamdani method and the Sugeno method.

[0070] In a specific implementation, in the fuzzy logic control algorithm, in A4, the defuzzification method is the centroid method and the maximum membership method.

[0071] In a specific implementation, in the fuzzy logic control algorithm, in A2, the fuzzy rule base is: if the pressure is large and the flow is small, then the opening of the drainage valve is large; if the pressure is medium and the flow is medium, then the opening of the drainage valve is medium; if the pressure is small and the flow is large, then the opening of the drainage valve is small.

[0072] In a specific implementation, the training process of the neural network training model is:

[0073] B1: Collect and organize data sets for training and testing, including intrathoracic air pressure, flow, patient age, weight, disease severity, and drainage valve opening, preprocess the data sets, and divide the data sets into training set, validation set, and test set;

[0074] B2: Network structure construction: Determine the number of layers and the number of neurons in each layer, including the number of input layers, hidden layers, and output layers and the number of neurons in each layer, and select activation functions such as Sigmoid, Tanh, ReLU, etc. Determine the loss function, such as mean square error (MSE) or cross-entropy loss (Cross-Entropy Loss).

[0075] B3: Initialize the parameters, then forward propagate, pass the input data to the input layer, hidden layer and output layer for calculation, calculate the activation function, and then calculate the difference between the output and the true value through the loss function, and then use the gradient descent method to update the parameters, repeat forward propagation, loss calculation, back propagation and parameter update until the predetermined number of iterations is reached or the loss function converges;

[0076] B4: Model evaluation, use the validation set to evaluate the performance of the model, adjust the hyperparameters to improve the model effect, and use the test set to evaluate the generalization ability of the model.

[0077] In a specific embodiment, a user interaction interface 3 is also included, which can provide an intuitive user interface, allowing medical staff to easily set and adjust treatment parameters, while also being able to monitor the patient's real-time data.

[0078] In a specific implementation, the user interaction interface 3 includes:

[0079] The display unit displays key parameters such as intrathoracic pressure and drainage flow in real time.

[0080] A control panel provides a manual control option, allowing medical staff to manually intervene when necessary.

[0081] The alarm system will automatically sound an alarm when it detects abnormal conditions, such as excessive pressure or abnormal flow.

[0082] In a specific implementation, the flow sensor is a differential pressure flow sensor, an ultrasonic flow sensor, or a thermal flow sensor, which can accurately measure the gas flow velocity.

[0083] The present invention also provides an adaptive pneumothorax pressure and flow intelligent control method, comprising the following steps:

[0084] S1: Installation and calibration of sensor module 1: Install the pressure sensor near the patient's chest to ensure that it can accurately monitor the intrathoracic pressure; install the flow sensor on the drainage tube 2 to ensure that it can accurately measure the gas flow; calibrate the sensor to ensure its measurement accuracy; insert the drainage tube 2 into the drainage tube placement 4;

[0085] S2: Configuration of the central processing module: configure the microprocessor to ensure that it can receive and process data from the sensor; configure the control algorithm unit, including the trained fuzzy logic control algorithm and neural network training model;

[0086] S3: To achieve closed-loop control, the sensor module 1 receives data in real time and inputs it into the control algorithm. The control algorithm calculates the optimal control signal and sends it to the actuator to adjust the pneumothorax pressure. The control signal is adjusted according to the real-time feedback to ensure that the pressure is maintained within the target range.

[0087] S4: User interaction interface 3, which displays intrathoracic pressure and drainage flow parameters, provides manual control options, and allows medical staff to manually intervene when necessary; implements an alarm system that automatically issues an alarm when an abnormal situation is detected.

[0088] Implement fault detection mechanisms so that problems can be detected and actions can be taken in a timely manner. Design emergency shutdown and manual control options to ensure patient safety and stable operation of equipment.

[0089] Components not described in detail in this embodiment are all existing components that can be purchased through public channels.

[0090] The above description of the embodiments is to facilitate the understanding and use of the invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. An adaptive pneumothorax pressure and flow intelligent control monitoring system, characterized in that: include: A sensor module (1), a central processing module, and an execution module; the central processing module is in communication connection with the sensor module (1) and the execution module; The sensor module (1) is arranged near the chest cavity of the patient, and comprises a pressure sensor and a flow sensor, wherein the pressure sensor is used to monitor the air pressure change in the chest cavity in real time, and the flow sensor is used to measure the gas flow through the drainage tube (2); the execution module is a drainage valve, and the execution module automatically adjusts the opening of the drainage valve according to the control signal of the central processing module to control the gas discharge speed; the central processing module comprises a microprocessor and a storage unit; The central processing module is provided with a fuzzy logic control algorithm and a neural network training model; The fuzzy logic control algorithm includes: A1: Fuzzification process: Fuzzification is performed on the input variables, which are pressure and flow, and a membership function is defined for each input variable to map the input value to the membership degree in the fuzzy set; A2: Establish a fuzzy rule base: Establish fuzzy rules to describe the relationship between input variables and output variables; the output variable is the opening of the drainage valve; A3: Fuzzy reasoning: Based on the membership degree of input variables and the fuzzy rule base, fuzzy reasoning is performed to obtain the fuzzy set of output variables; A4: Defuzzification: convert the fuzzy set of output variables obtained by fuzzy reasoning into specific numerical output, i.e. the opening of the diversion valve; The neural network training model is a feedforward neural network model, which includes an input layer, a hidden layer and an output layer; the pressure and flow change data and the patient's age, weight, and severity of the disease are input into the input layer, the output layer outputs the data for the prediction of the drainage valve opening, the hidden layer extracts complex features from the input data, and introduces nonlinearity through an activation function to achieve prediction; The fuzzy logic control algorithm and the neural network training model are combined to perform data processing analysis and prediction, conduct real-time monitoring, and control the opening of the drainage valve.

2. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: In the fuzzy logic control algorithm, in A1, the membership function is a triangular function, a trapezoidal function or a Gaussian function.

3. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: In the fuzzy logic control algorithm, in A3, the fuzzy reasoning method is the Mamdani method and the Sugeno method.

4. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: In the fuzzy logic control algorithm, in A4, the defuzzification method is the centroid method and the maximum membership method.

5. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: In the fuzzy logic control algorithm, in A2, the fuzzy rule base is: if the pressure is large and the flow is small, then the opening of the drainage valve is large; if the pressure is medium and the flow is medium, then the opening of the drainage valve is medium; If the pressure is low and the flow is high, the diversion valve is open a small amount.

6. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: The training process of the neural network training model is: B1: Collect and organize data sets for training and testing, including intrathoracic air pressure, flow, patient age, weight, disease severity, and drainage valve opening, preprocess the data sets, and divide the data sets into training set, validation set, and test set; B2: Network structure construction: determine the number of layers and the number of neurons in each layer, including the number of input layers, hidden layers and output layers and the number of neurons in each layer, select the activation function, and determine the loss function; B3: Initialize the parameters, then forward propagate, pass the input data to the input layer, hidden layer and output layer for calculation, calculate the activation function, and then calculate the difference between the output and the true value through the loss function, and then use the gradient descent method to update the parameters, repeat forward propagation, loss calculation, back propagation and parameter update until the predetermined number of iterations is reached or the loss function converges; B4: Model evaluation, use the validation set to evaluate the performance of the model, adjust the hyperparameters to improve the model effect, and use the test set to evaluate the generalization ability of the model.

7. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: It also includes a user interaction interface (3) that can provide an intuitive user interface, allowing medical staff to easily set and adjust treatment parameters while also being able to monitor real-time data of the patient.

8. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 7, characterized in that: The user interaction interface (3) comprises: Display unit, which displays key parameters such as intrathoracic pressure and drainage flow in real time; A control panel that provides manual control options, allowing medical staff to manually intervene when necessary; The alarm system will automatically sound an alarm when it detects abnormal conditions, such as excessive pressure or abnormal flow.

9. The adaptive pneumothorax pressure and flow intelligent control monitoring system according to claim 1, characterized in that: The flow sensor is a differential pressure flow sensor, an ultrasonic flow sensor or a thermal flow sensor, which can accurately measure the gas flow velocity.

10. An adaptive pneumothorax pressure and flow intelligent control method of the system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1: Installation and calibration of the sensor module (1): Install the pressure sensor near the patient's chest to ensure that it can accurately monitor the intrathoracic pressure; install the flow sensor on the drainage tube (2) to ensure that it can accurately measure the gas flow; calibrate the sensor to ensure its measurement accuracy; insert the drainage tube (2) into the drainage tube placement (4); S2: Configuration of the central processing module: configure the microprocessor to ensure that it can receive and process data from the sensor; configure the control algorithm unit, including the trained fuzzy logic control algorithm and neural network training model; S3: Implementing closed-loop control, the sensor module (1) receives data in real time and inputs it into the control algorithm, the control algorithm calculates the optimal control signal and sends it to the actuator to adjust the pneumothorax pressure, and adjusts the control signal according to the real-time feedback to ensure that the pressure is maintained within the target range; S4: User interaction interface (3), which displays intrathoracic pressure and drainage flow parameters, provides manual control options, and allows medical staff to manually intervene when necessary; implements an alarm system that automatically issues an alarm when an abnormal situation is detected.

Images