Mammary gland postoperative external chest fixation pressure monitoring and early warning method

Abstract

The invention discloses a mammary gland postoperative external chest fixation pressure monitoring and early warning method, which relates to the technical field of equipment monitoring and comprises the following steps: S1, postoperative evaluation and equipment selection; s2, arranging and calibrating a pressure sensor; s3, performing multi-dimensional acquisition on pressure data; s4, high-frequency data acquisition and compression transmission; s5, performing dynamic analysis and anomaly detection on the pressure data; s6, constructing a personalized pressure early warning model based on AI; and S7, designing a multi-stage pressure early warning mechanism. According to the early warning method, intelligent pressure monitoring and early warning related processes are introduced, real-time and automatic monitoring and dynamic analysis can be carried out on the pressure applied to the external chest fixing device, and therefore potential abnormal conditions can be recognized in time; according to the method, the individual physiological features of the patient and the postoperative recovery condition are combined, the early warning threshold value can be intelligently adjusted, the monitoring accuracy and the individuation level are improved, then complications are effectively reduced, and the postoperative recovery experience of the patient is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and in particular to a method for monitoring and early warning of chest external fixation pressure after breast surgery. Background Art

[0002] After breast surgery, patients often need to use external chest fixation devices to support the postoperative area and prevent complications. However, existing monitoring technologies mainly rely on regular manual inspections and cannot monitor chest pressure changes in real time. This method not only has the limitations of subjective judgment, but may also lead to delayed responses to abnormal patient conditions. In addition, traditional monitoring methods are difficult to take into account the individual differences of each patient, and cannot achieve personalized pressure warnings, which can easily lead to false alarms or missed alarms, increasing the risk to patients.

[0003] In order to solve the above problems, we proposed a postoperative chest fixation pressure monitoring and early warning method for breast surgery. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for monitoring and early warning of chest fixation pressure after breast surgery to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for monitoring and early warning of chest external fixation pressure after breast surgery, comprising the following steps: S1, postoperative evaluation and equipment selection; Conduct a comprehensive assessment of the patient's overall chest condition after surgery and determine the appropriate external chest fixation device and pressure sensor; S2. Layout and calibration of pressure sensors; The sensors are distributed at key pressure points on the chest, such as near the wound, under the armpit and on the edge of the breast. The placement of the sensors is based on the patient's physical characteristics and the surgical area to accurately sense the pressure at each key part. After the placement is completed, the pressure sensors are calibrated. S3, multi-dimensional collection of pressure data; The multi-dimensional data collection mode is used to collect the pressure data of the chest fixation device on different parts. The multi-dimensional data includes real-time pressure, cumulative pressure and dynamic pressure changes. S4, high-frequency data acquisition and compression transmission; Conduct high-frequency real-time monitoring and transmit pressure data to the central monitoring system or mobile terminal through data compression technology; S5. Dynamic analysis and anomaly detection of pressure data; After receiving the pressure data, the system analyzes the data through cloud or local models; based on the historical data model and the current pressure change trend, detect whether there are potential anomalies; S6. Build an AI-based personalized stress warning model; Through AI technology, a personalized stress warning model is established according to each patient's postoperative recovery. The model automatically adjusts the sensitivity of the stress warning based on historical data, the patient's recovery speed and individual physiological characteristics. S7. Design a multi-level stress warning mechanism; The stress warning is designed as a multi-level mechanism, including mild, moderate and severe alarms.

[0006] To further optimize the technical solution, in step S1, the postoperative evaluation includes the wound healing, the sensitivity of the breast surgery area, and the possibility of swelling or effusion; the external chest fixation device adjusts the support position according to the patient's specific chest shape, wound location and physical condition; and the pressure sensor is embedded in the external chest fixation device.

[0007] To further optimize the technical solution, in step S2, when calibrating the pressure sensor, the following process is specifically included: Setting standard pressure points: The standard pressure points are determined by clinical experience and the individual conditions of postoperative patients, including the normal range of pressure applied by the chest fixation device to the wound, armpit, and breast edge. The standard pressure points are set based on historical data or standard medical reference values ​​according to the needs of different patients. Simulated pressure test: Use external equipment such as a standard pressure gauge to simulate standard pressures at different locations, and gradually apply known pressure values ​​to the sensors in the chest fixation device to provide known reference values ​​for the sensors as a benchmark for subsequent calibration. Adjust sensor sensitivity: adjust the sensitivity through gain adjustment and zero offset correction; Dynamic calibration and pressure change response test: By adjusting the patient's body position to different positions, such as supine and side-lying, common postoperative activity scenarios are simulated to monitor the response speed and accuracy of the sensor; Calibration data recording and model correction: Based on the calibration data, a personalized calibration model is formed for each patient, and automatic correction is performed based on the calibration data to improve the working efficiency and accuracy of the sensor; Regular recalibration and real-time calibration: Perform regular recalibration or automatically perform real-time calibration during monitoring through an adaptive algorithm based on the patient's recovery progress.

[0008] To further optimize the technical solution, in step S3, the real-time pressure is the pressure value at the current moment, the accumulated pressure is the total pressure in a fixed time period, and the dynamic pressure change shows the fluctuation of pressure over time; Based on multi-dimensional data collection, it is used to fully understand the stress conditions of different positions of the chest under the action of fixation devices.

[0009] To further optimize the technical solution, in step S4, the sensor collects pressure data hundreds of times per second, and preliminarily processes the pressure data through edge computing technology, filters invalid data, retains key change information, and uses a data compression algorithm to reduce the amount of transmitted data, thereby ensuring that the pressure data is transmitted to the central monitoring system or mobile terminal in real time without affecting the accuracy of the data.

[0010] To further optimize the technical solution, in step S5, the cloud or local model uses a machine learning model to perform real-time dynamic evaluation of the pressure data, and the machine learning model includes the following process: A supervised learning-based model is used to detect anomalies in pressure data using the Random Forest algorithm. Potential anomalies are identified based on historical pressure data and current real-time pressure values. In the model, multiple features are first extracted from the pressure sensor as input to the model. The features include: : Current time point The pressure value; :past The average pressure value at each time point; :past The maximum pressure value at a time point; :past The minimum pressure value at a time point; : The difference between the current pressure value and the average pressure value in the past ; : The duration of the current pressure value, such as the duration that it remains above the threshold; The feature extraction formula is: ; Use the decision tree ensemble learning method for classification and set a threshold judgment formula: ; in, and It is an adjustable hyperparameter that controls the sensitivity and duration of the warning; Indicates the ratio threshold of the current pressure to the average pressure. For example, if it is set to 1.2, the alarm is triggered when the current pressure is higher than 120% of the past average pressure; Indicates the time limit for the pressure to remain above the threshold. For example, if it is set to 5 minutes, an alarm will be issued only if the pressure remains above the threshold for 5 minutes.

[0011] To further optimize the technical solution, in step S6, building a stress warning model includes the following process: Build a deep neural network (DNN) model to achieve personalized learning and early warning of pressure data. The input of the model includes the patient's postoperative historical data, real-time pressure data and physiological characteristics; The input features of the model are defined as: : The pressure value at the current time point; :past The average pressure value at each time point; : The patient's recovery speed, such as wound healing index; : The patient's physical characteristics, such as age, gender, and weight; : Patients’ psychological status scores, such as pain level and anxiety level; The feature input formula is: ; Use a multi-layer perceptron (MLP) to output personalized stress warning thresholds , the basic form of the model is as follows: ; in, is an activation function, such as ReLU or Sigmoid, used to introduce nonlinearity; is a weight matrix that learns the importance of different features through training; is the bias term, which adjusts the model output; Personalized thresholds via model output , combined with the safety range and individual characteristics in historical data, the specific implementation of the early warning is finally determined: .

[0012] To further optimize the technical solution, in step S7, a minor alarm is triggered when the pressure approaches a preset safety threshold, prompting the patient or medical staff to pay attention; a moderate alarm is issued when the pressure slightly exceeds the threshold, reminding the patient to make appropriate adjustments; a severe alarm is triggered when the pressure greatly exceeds the safety range, immediately notifying the medical staff or emergency treatment; The pressure warning mechanism transmits information in real time through vibration, sound, and APP push to ensure timely response.

[0013] To further optimize this technical solution, the method also uploads all pressure data to the cloud, and doctors or nursing staff can view the patient's pressure changes in real time through the central monitoring system or mobile terminal; The central monitoring system or mobile terminal not only displays real-time pressure data, but also generates trend reports based on the historical data of different patients to help medical staff make scientific postoperative care decisions. Patients can also view their own pressure data in real time through the mobile phone APP to understand the status of the external chest fixation device.

[0014] To further optimize the technical solution, the method also regularly conducts feedback analysis on the collected data, and optimizes the pressure monitoring algorithm and warning threshold based on the patient's postoperative recovery progress; For example, regularly check the frequency and accuracy of pressure alarms, correct unnecessary alarms, and improve the accuracy of early warnings; Through feedback data, the sensitivity of pressure monitoring is automatically adjusted to reduce the risk of too frequent warnings or missed reports. Long-term data accumulation is also used as a reference for future patients' postoperative recovery.

[0015] Compared with the prior art, the present invention provides a method for monitoring and early warning of chest external fixation pressure after breast surgery, which has the following beneficial effects: This postoperative chest fixation pressure monitoring and early warning method for breast surgery can achieve real-time, automatic monitoring and dynamic analysis of the pressure applied by the chest fixation device by introducing intelligent pressure monitoring and early warning related processes, so as to timely identify potential abnormal conditions; this method combines the patient's individual physiological characteristics and postoperative recovery status, and can intelligently adjust the early warning threshold, improve the accuracy and personalization level of monitoring, thereby effectively reducing the occurrence of complications and improving the patient's postoperative recovery experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a method for monitoring and early warning of chest external fixation pressure after breast surgery proposed by the present invention; Figure 2 A schematic diagram of the process of calibrating a pressure sensor in a breast surgery external fixation pressure monitoring and early warning method proposed by the present invention; Figure 3 A flow chart of a machine learning model in a breast surgery external chest fixation pressure monitoring and early warning method proposed by the present invention; Figure 4 The present invention is a flow chart of a pressure warning model in a breast surgery external chest fixation pressure monitoring and warning method proposed by the present invention. DETAILED DESCRIPTION

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

[0018] See also Figure 1 , a breast surgery chest fixation pressure monitoring and early warning method, comprising the following steps: S1. Postoperative evaluation and equipment selection Perform a comprehensive assessment of the patient's overall chest condition after surgery and determine the appropriate external chest fixation device and pressure sensor.

[0019] In this embodiment, in step S1, the postoperative evaluation includes the wound healing, the sensitivity of the breast surgery area, and the possibility of swelling or effusion. The application of the external chest fixation device must be adjusted according to the patient's specific chest morphology, wound location and physical condition to ensure that the external fixation device does not cause excessive pressure on the wound or hinder wound healing. The device should have good air permeability, elasticity and adjustability, and can provide personalized support and compression for different patients. The material and design of the fixation device also need to consider the special needs of postoperative recovery, such as avoiding excessive compression of blood vessels or causing secondary injuries. The pressure sensor is embedded in the interior of the external chest fixation device, and the pressure sensor can monitor the pressure between the fixation device and the patient's skin surface in real time. The pressure sensor needs to have high sensitivity and durability, can work uninterruptedly for a long time and have anti-interference ability to ensure the stability and accuracy of the data. During installation, special attention should be paid to avoid direct contact between the pressure sensor and the wound or allergic skin to prevent discomfort or other skin problems.

[0020] S2. Layout and calibration of pressure sensors The sensors are distributed at key pressure points on the chest, such as near the wound, under the armpits and on the edge of the breast. The sensors are placed based on the patient's physical characteristics and surgical area to accurately sense the pressure of each key part. After the placement is completed, the pressure sensors are calibrated.

[0021] like Figure 2 As shown, in this embodiment, in step S2, when calibrating the pressure sensor, the following process is specifically included: Setting standard pressure points: Standard pressure points are determined by clinical experience and the individual conditions of postoperative patients, including the normal range of pressure applied by external chest fixation devices to the wound, armpits, and breast edges. Standard pressure points are set based on historical data or standard medical reference values ​​according to the needs of different patients.

[0022] Simulated pressure test: Use external equipment such as a standard pressure device to simulate standard pressures at different parts of the body, and gradually apply known pressure values ​​to the sensors in the external chest fixation device to provide known reference values ​​for the sensors, which serve as a benchmark for subsequent calibration. Manual pressure adjustment refers to the operator applying different intensities of pressure (such as light pressure, medium pressure, and strong pressure) to the fixation device manually or mechanically, and then recording the data returned by the sensor to confirm whether it is consistent with the set standard pressure. Multi-point pressure distribution means that the calibration of each sensor should be performed independently, testing the pressure response of each key part one by one, and ensuring that the sensor can accurately respond to pressure changes at different locations.

[0023] Adjust sensor sensitivity: adjust the sensitivity through gain adjustment and zero offset correction. Gain adjustment refers to adjusting the gain parameter of the sensor to make it more sensitive to small pressure changes. For example, if the sensor does not respond well to small pressure changes, the gain needs to be increased so that it can recognize smaller pressure changes. Zero offset correction means that in the absence of external pressure, the sensor may have a zero offset, that is, the sensor displays a non-zero value in the no-pressure state. At this time, zero calibration is required to ensure that the value output by the sensor is zero when there is no pressure, to avoid errors in the subsequent data collection process.

[0024] Dynamic calibration and pressure change response test: By adjusting the patient's body position to different positions, such as lying flat or lying on the side, simulating common postoperative activity scenarios, the sensor's response speed and accuracy are monitored. The response time test refers to checking the time from the sensor's pressure change to the output signal to ensure that it can respond quickly in a short time. The pressure change linearity test refers to monitoring whether the sensor's output data is linearly related to the pressure change by applying gradually increasing or decreasing pressure to ensure that the device can accurately capture the pressure change process.

[0025] Calibration data recording and model correction: A personalized calibration model for each patient is formed based on the calibration data, and automatic correction is performed based on the calibration data to improve the working efficiency and accuracy of the sensor.

[0026] Regular recalibration and real-time calibration: Perform regular recalibration or automatically perform real-time calibration during monitoring through an adaptive algorithm based on the patient's recovery progress.

[0027] Through a sophisticated calibration process, the sensor can effectively sense the pressure of the chest fixation device on the patient's chest, ensuring the accuracy of the monitoring data. This meticulous calibration not only improves the accuracy and stability of the system, but also provides a reliable basis for subsequent pressure warnings.

[0028] S3. Multi-dimensional collection of pressure data A multi-dimensional data acquisition mode is used to collect pressure data of different parts of the chest fixation device. The multi-dimensional data includes real-time pressure, cumulative pressure and dynamic pressure changes.

[0029] In this embodiment, in step S3, the real-time pressure is the pressure value at the current moment, the accumulated pressure is the total pressure value within a fixed time period, and the dynamic pressure change shows the fluctuation of pressure over time; Based on multi-dimensional data collection, it is used to fully understand the stress conditions of different positions of the chest under the action of fixation devices.

[0030] S4, high frequency data acquisition and compression transmission; Conduct high-frequency real-time monitoring, and process it through data compression technology to transmit the pressure data to the central monitoring system or mobile terminal.

[0031] In this embodiment, in step S4, the sensor collects pressure data hundreds of times per second, and preliminarily processes the pressure data through edge computing technology, filters invalid data, retains key change information, and uses a data compression algorithm to reduce the amount of transmitted data, thereby ensuring that the pressure data is transmitted to the central monitoring system or mobile terminal in real time without affecting the accuracy of the data.

[0032] S5. Dynamic analysis and anomaly detection of pressure data After receiving the pressure data, the system analyzes the data through cloud or local models; based on the historical data model and the current pressure change trend, it detects whether there are potential anomalies.

[0033] like Figure 3 As shown, in this embodiment, in step S5, the cloud or local model uses a machine learning model to perform real-time dynamic evaluation of the pressure data, and the machine learning model includes the following process: A supervised learning-based model is used to detect anomalies in pressure data using the Random Forest algorithm. Potential anomalies are identified based on historical pressure data and current real-time pressure values. In the model, multiple features are first extracted from the pressure sensor as input to the model. The features include: : Current time point The pressure value; :past The average pressure value at each time point; :past The maximum pressure value at a time point; :past The minimum pressure value at a time point; : The difference between the current pressure value and the average pressure value in the past ; : The duration of the current pressure value, such as the duration that it remains above the threshold; The feature extraction formula is: ; Use the decision tree ensemble learning method for classification and set a threshold judgment formula: ; in, and It is an adjustable hyperparameter that controls the sensitivity and duration of the warning; Indicates the ratio threshold of the current pressure to the average pressure. For example, if it is set to 1.2, the alarm is triggered when the current pressure is higher than 120% of the past average pressure; Indicates the time limit for the pressure to remain above the threshold. For example, if it is set to 5 minutes, an alarm will be issued only if the pressure remains above the threshold for 5 minutes.

[0034] Combining the above formulas and features, the model usage process is as follows: Data Collection The system collects pressure data in real time and extracts the required features. It updates the current pressure value and the past pressure values ​​at regular intervals (e.g., every minute). Pressure data at a certain time point.

[0035] Feature calculation Calculate required features based on real-time data: Update current pressure value ; Calculate the past The average pressure at each time point , Maximum pressure and minimum pressure ; Calculate pressure change ; Calculate the current pressure duration .

[0036] Model Inference The extracted features are input into the random forest model for classification and anomaly detection using the formula: If the model returns , the system triggers an alarm to remind medical staff to conduct further examination; If it returns , the system continues to monitor without taking any action.

[0037] Model Optimization The system will continue to collect feedback data, record each alarm situation, and retrain the model to improve future detection accuracy. It can be optimized through methods such as cross-validation. and parameters to ensure a balance between the sensitivity and specificity of the model.

[0038] By building such a machine learning model, the system can monitor the pressure changes of the chest fixation device in real time and quickly identify potential abnormalities. This anomaly detection model based on random forests, combined with specific formulas and features, provides an efficient and accurate pressure monitoring and early warning mechanism, which can greatly improve the safety and effectiveness of postoperative care.

[0039] S6. Build a personalized stress warning model based on AI Through AI technology, a personalized stress warning model is established according to the postoperative recovery of each patient. The model automatically adjusts the sensitivity of the stress warning based on historical data, the patient's recovery speed and individual physiological characteristics.

[0040] like Figure 4 As shown, in this embodiment, in step S6, building a stress warning model includes the following process: Build a deep neural network (DNN) model to achieve personalized learning and early warning of pressure data. The input of the model includes the patient's postoperative historical data, real-time pressure data and physiological characteristics; The input features of the model are defined as: : The pressure value at the current time point; :past The average pressure value at each time point; : The patient's recovery speed, such as wound healing index; : The patient's physical characteristics, such as age, gender, and weight; : Patients’ psychological status scores, such as pain level and anxiety level; The feature input formula is: ; Use a multi-layer perceptron (MLP) to output personalized stress warning thresholds , the basic form of the model is as follows: ; in, is an activation function, such as ReLU or Sigmoid, used to introduce nonlinearity; is a weight matrix that learns the importance of different features through training; is the bias term, which adjusts the model output; Personalized thresholds via model output , combined with the safety range and individual characteristics in historical data, the specific implementation of the early warning is finally determined: .

[0041] Combining the above formulas and steps, the model usage process is as follows: Data collection and feature extraction The system collects the patient's pressure data and physiological characteristics in real time and extracts input features The feature extraction includes the current stress value, the average value of past stress, the patient's recovery speed, basic physiological data, and psychological status score.

[0042] Model Inference Input the extracted features into the trained deep learning model to calculate the personalized warning threshold At this point, the model adjusts the weights based on historical data and patient characteristics to optimize the output.

[0043] Generate alert Compare current pressure values and output personalized thresholds : if , the system will sound an alarm, prompting medical staff to evaluate and handle; if , the system continues to monitor without processing.

[0044] Model self-learning and optimization The system should regularly update and train the model, adjusting weights based on new pressure monitoring data and patient feedback and bias . Incremental learning can be used to regularly introduce new data for model retraining to improve its accuracy and personalization capabilities.

[0045] Patient feedback and adaptation Through patient feedback (such as comfort surveys), combined with psychological status scores , the model can continuously adjust itself to optimize the sensitivity of the warning threshold. For example, if the patient is in a good mental state and recovers quickly, the stress threshold can be appropriately increased, thereby reducing the frequency of unnecessary alarms.

[0046] By building an AI-based personalized stress warning model, the system can adapt to the specific situation of each patient in real time, improving the accuracy and personalization of monitoring. This deep learning model can not only respond sensitively in the early postoperative period, but also has the ability to dynamically adjust in the later stages of recovery, providing safer and more effective support for postoperative care.

[0047] S7. Design a multi-level pressure warning mechanism The stress warning is designed as a multi-level mechanism, including mild, moderate and severe alarms.

[0048] In the present embodiment, in step S7, a minor alarm is triggered when the pressure approaches a preset safety threshold, prompting the patient or medical staff to pay attention; a moderate alarm is issued when the pressure slightly exceeds the threshold, reminding the patient to make appropriate adjustments; a severe alarm is triggered when the pressure greatly exceeds the safety range, immediately notifying the medical staff or emergency treatment; The pressure warning mechanism transmits information in real time through vibration, sound, and APP push to ensure timely response.

[0049] In this embodiment, the method also uploads all pressure data to the cloud, and doctors or nursing staff can view the patient's pressure changes in real time through a central monitoring system or a mobile terminal; The central monitoring system or mobile terminal not only displays real-time pressure data, but also generates trend reports based on the historical data of different patients to help medical staff make scientific postoperative care decisions. Patients can also view their own pressure data in real time through the mobile phone APP to understand the status of the external chest fixation device.

[0050] At the same time, the method also regularly conducts feedback analysis on the collected data, and optimizes the pressure monitoring algorithm and warning threshold based on the patient's postoperative recovery progress; For example, regularly check the frequency and accuracy of pressure alarms, correct unnecessary alarms, and improve the accuracy of early warnings; Through feedback data, the sensitivity of pressure monitoring is automatically adjusted to reduce the risk of too frequent warnings or missed reports. Long-term data accumulation is also used as a reference for future patients' postoperative recovery.

[0051] The beneficial effects of the present invention are: This postoperative chest fixation pressure monitoring and early warning method for breast surgery can achieve real-time, automatic monitoring and dynamic analysis of the pressure applied by the chest fixation device by introducing intelligent pressure monitoring and early warning related processes, so as to timely identify potential abnormal conditions; this method combines the patient's individual physiological characteristics and postoperative recovery status, and can intelligently adjust the early warning threshold, improve the accuracy and personalization level of monitoring, thereby effectively reducing the occurrence of complications and improving the patient's postoperative recovery experience.

[0052] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring and warning postoperative chest fixation pressure in breast surgery. It is characterized in that The following steps are involved: S1, postoperative evaluation and equipment selection; Conduct a comprehensive assessment of the patient's overall chest condition after surgery and determine the appropriate external chest fixation device and pressure sensor; S2. Layout and calibration of pressure sensors; Sensors are distributed at key pressure points on the chest, such as near the wound, under the armpits and on the edge of the breast. The placement of sensors is based on the patient's physical characteristics and surgical area to accurately sense the pressure at each key part. After the deployment is completed, the pressure sensor is calibrated; S3, multi-dimensional collection of pressure data; The multi-dimensional data collection mode is used to collect the pressure data of the chest fixation device on different parts. The multi-dimensional data includes real-time pressure, cumulative pressure and dynamic pressure changes. S4, high-frequency data acquisition and compression transmission; Conduct high-frequency real-time monitoring and transmit pressure data to the central monitoring system or mobile terminal through data compression technology; S5. Dynamic analysis and anomaly detection of pressure data; After receiving the pressure data, the system analyzes the data through cloud or local models; based on the historical data model and the current pressure change trend, detect whether there are potential anomalies; S6. Build an AI-based personalized stress warning model; Through AI technology, a personalized stress warning model is established according to each patient's postoperative recovery. The model automatically adjusts the sensitivity of the stress warning based on historical data, the patient's recovery speed and individual physiological characteristics. S7. Design a multi-level stress warning mechanism; The stress warning is designed as a multi-level mechanism, including mild, moderate and severe alarms.

2. A method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S1, the postoperative evaluation includes the wound healing, the sensitivity of the breast surgery area, and the possibility of swelling or fluid accumulation; the external chest fixation device adjusts the support position according to the patient's specific chest shape, wound location and physical condition; and the pressure sensor is embedded in the internal part of the external chest fixation device.

3. A method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S2, when calibrating the pressure sensor, the following process is specifically included: Setting standard pressure points: The standard pressure points are determined by clinical experience and the individual conditions of postoperative patients, including the normal range of pressure applied by the chest fixation device to the wound, armpit, and breast edge. The standard pressure points are set based on historical data or standard medical reference values ​​according to the needs of different patients. Simulated pressure test: Use external equipment such as a standard pressure gauge to simulate standard pressures at different locations, and gradually apply known pressure values ​​to the sensors in the chest fixation device to provide known reference values ​​for the sensors as a benchmark for subsequent calibration. Adjust sensor sensitivity: adjust the sensitivity through gain adjustment and zero offset correction; Dynamic calibration and pressure change response test: By adjusting the patient's body position to different positions, such as supine and side-lying, common postoperative activity scenarios are simulated to monitor the response speed and accuracy of the sensor; Calibration data recording and model correction: Based on the calibration data, a personalized calibration model is formed for each patient, and automatic correction is performed based on the calibration data to improve the working efficiency and accuracy of the sensor; Regular recalibration and real-time calibration: Perform regular recalibration or automatically perform real-time calibration during monitoring through an adaptive algorithm based on the patient's recovery progress.

4. A method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S3, the real-time pressure is the pressure value at the current moment, the accumulated pressure is the total pressure in a fixed time period, and the dynamic pressure change shows the fluctuation of pressure over time; Based on multi-dimensional data collection, it is used to fully understand the stress conditions of different positions of the chest under the action of fixation devices.

5. The method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S4, the sensor collects pressure data hundreds of times per second, and preliminarily processes the pressure data through edge computing technology, filters invalid data, retains key change information, and uses data compression algorithm to reduce the amount of transmitted data, ensuring that the pressure data is transmitted to the central monitoring system or mobile terminal in real time without affecting the accuracy of the data.

6. A method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S5, the cloud or local model uses a machine learning model to perform real-time dynamic evaluation of the pressure data. The machine learning model includes the following process: A supervised learning-based model is used to detect anomalies in pressure data using the Random Forest algorithm. Potential anomalies are identified based on historical pressure data and current real-time pressure values. In the model, multiple features are first extracted from the pressure sensor as input to the model. The features include: : Current time point The pressure value; :past The average pressure value at each time point; :past The maximum pressure value at a time point; :past The minimum pressure value at a time point; : The difference between the current pressure value and the average pressure value in the past ; : The duration of the current pressure value, such as the duration that it remains above the threshold; The feature extraction formula is: ； Use the decision tree ensemble learning method for classification and set a threshold judgment formula: ； in, and It is an adjustable hyperparameter that controls the sensitivity and duration of the warning; Indicates the ratio threshold of the current pressure to the average pressure. For example, if it is set to 1.2, the alarm is triggered when the current pressure is higher than 120% of the past average pressure; Indicates the time limit for the pressure to remain above the threshold. For example, if it is set to 5 minutes, an alarm will be issued only if the pressure remains above the threshold for 5 minutes.

7. A method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S6, constructing a stress warning model includes the following process: Build a deep neural network (DNN) model to achieve personalized learning and early warning of pressure data. The input of the model includes the patient's postoperative historical data, real-time pressure data and physiological characteristics; The input features of the model are defined as: : The pressure value at the current time point; :past The average pressure value at each time point; : The patient's recovery speed, such as wound healing index; : The patient's physical characteristics, such as age, gender, and weight; : Patients’ psychological status scores, such as pain level and anxiety level; The feature input formula is: ； Use a multi-layer perceptron (MLP) to output personalized stress warning thresholds , the basic form of the model is as follows: ； in, is an activation function, such as ReLU or Sigmoid, used to introduce nonlinearity; is a weight matrix that learns the importance of different features through training; is the bias term, which adjusts the model output; Personalized thresholds via model output , combined with the safety range and individual characteristics in historical data, the specific implementation of the early warning is finally determined: 。 8. The method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that In step S7, a minor alarm is triggered when the pressure approaches a preset safety threshold, prompting the patient or medical staff to pay attention; a moderate alarm is issued when the pressure slightly exceeds the threshold, reminding the patient to make appropriate adjustments; a severe alarm is triggered when the pressure greatly exceeds the safety range, immediately notifying the medical staff or emergency treatment; The pressure warning mechanism transmits information in real time through vibration, sound, and APP push to ensure timely response.

9. The method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that This method also uploads all pressure data to the cloud, so that doctors or nursing staff can view the patient's pressure changes in real time through the central monitoring system or mobile terminal; The central monitoring system or mobile terminal not only displays real-time pressure data, but also generates trend reports based on the historical data of different patients to help medical staff make scientific postoperative care decisions. Patients can also view their own pressure data in real time through the mobile phone APP to understand the status of the external chest fixation device.

10. The method for monitoring and early warning of chest external fixation pressure after breast surgery according to claim 1, It is characterized in that The method also regularly conducts feedback analysis on the collected data and optimizes the pressure monitoring algorithm and warning threshold based on the patient's postoperative recovery progress; For example, regularly check the frequency and accuracy of pressure alarms, correct unnecessary alarms, and improve the accuracy of early warnings; Through feedback data, the sensitivity of pressure monitoring is automatically adjusted to reduce the risk of too frequent warnings or missed reports. Long-term data accumulation is also used as a reference for future patients' postoperative recovery.

Images