Intelligent gastrointestinal decompression device based on sensor and control method

Abstract

The invention discloses an intelligent gastrointestinal decompression device based on a sensor and a control method, and belongs to the field of medical equipment. The invention discloses an intelligent gastrointestinal decompression device based on a sensor and a control method. The intelligent gastrointestinal decompression device comprises a data acquisition unit, an intelligent control unit and an intelligent protection unit. The problem that personalized adjustment cannot be carried out according to specific conditions of patients in the prior art is solved, various parameters in the stomach are monitored in real time through the integrated sensor array, data accuracy can be ensured by periodically calibrating the sensor, a gastrointestinal state model is constructed based on deep learning, and the accuracy of the gastrointestinal state is improved. The intelligent gastrointestinal decompression device accurately evaluates the illness state, generates a personalized control instruction, adopts a self-adaptive closed-loop control strategy, can guarantee the operation accuracy and stability, predicts the risk in real time, automatically switches to a safety mode, generates different levels of early warning signals according to the risk probability, and improves the safety of the intelligent gastrointestinal decompression device. And medical workers are prompted to perform targeted treatment in time.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment, and in particular to a sensor-based intelligent gastrointestinal decompression device and a control method. Background Art

[0002] Gastrointestinal decompression is a common clinical treatment method, mainly used to relieve pressure in the gastrointestinal tract, reduce patient discomfort, and promote the recovery of gastrointestinal function. Traditional gastrointestinal decompression devices mainly use negative pressure suction to expel gas and liquid from the gastrointestinal tract. Existing gastrointestinal decompression devices usually use fixed negative pressure values ​​and working modes, and cannot be personalized according to the specific conditions of the patients, making it difficult to meet the needs of different patients in different physiological and pathological conditions; therefore, it does not meet the existing needs. In this regard, we propose a sensor-based intelligent gastrointestinal decompression device and control method. Summary of the Invention

[0003] The purpose of the present invention is to provide a sensor-based intelligent gastrointestinal decompression device and control method. By integrating a sensor array in the gastric tube, the gastric pressure, content properties, pH and temperature are monitored in real time. By fusing multi-source data to build a gastrointestinal state model, personalized control instructions can be dynamically generated. Then, an adaptive closed-loop control strategy is adopted to dynamically adjust the negative pressure value, working mode and duration, thereby realizing the control of the intelligent gastrointestinal decompression device. At the same time, an active protection mechanism is built in to predict the risk of aspiration, mucosal damage or blockage based on real-time data, automatically switch to a safe mode and issue an early warning, thereby solving the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned object, the present invention provides the following technical solutions: a sensor-based intelligent gastrointestinal decompression device, the intelligent gastrointestinal decompression device comprising a data acquisition unit, an intelligent control unit and an intelligent protection unit; The data acquisition unit is configured to be integrated with a sensor array on the gastric tube, and the sensor array is used to monitor the intragastric pressure, content properties, pH and temperature in real time; Based on the data collected in real time by the data acquisition unit, the risk of aspiration, mucosal injury and blockage is predicted in real time, providing a more physiological and pathological risk assessment result for real-time clinical intervention. If there is a risk, a corresponding risk signal is generated to alert medical staff. The intelligent control unit is configured to receive data collected by the sensor array, fuse multi-source data to construct a gastrointestinal state model, dynamically generate personalized control instructions, and then use an adaptive closed-loop control strategy to dynamically adjust the negative pressure value, working mode and working time of the intelligent gastrointestinal decompression device; The intelligent protection unit is configured as the intelligent gastrointestinal decompression device with a built-in active protection mechanism for predicting the risk of aspiration, mucosal damage or blockage based on real-time data, and automatically switching to a safe mode while triggering an early warning.

[0005] Furthermore, the data acquisition unit includes: a data calibration module configured to periodically calibrate the sensor array and perform a verification test after calibration to check whether the sensor output is within an allowable error range by repeatedly applying a known standard value. If the error range is exceeded, a reminder is given to recalibrate and the number of calibration failures is recorded. If the number of recorded failures exceeds a preset number, it is indicated that the sensor is faulty or damaged; The calibration process is as follows: the data calibration module sends a calibration instruction to the sensor array, and the sensor starts collecting data and sends it back to the data calibration module; The data calibration module compares the data collected by the sensor with the known standard value and analyzes the deviation between the data collected by the sensor and the known standard value; Based on the deviation analysis results, the data calibration module adjusts the calibration parameters of the sensor array, including correcting the sensitivity and offset parameters of the sensors; The data processing module is configured to pre-process the data collected by the sensor array, including: Through data cleaning, outliers, erroneous data and redundant data are removed from the sensor collected data; Adopting adaptive filtering algorithm to remove high-frequency noise and low-frequency drift in sensor collected data in real time; A microprocessor is used to perform preliminary operations on the data, converting the raw data into physiological parameter values, and encoding and packaging the data using a coding format to package the data into a data frame. Each data frame contains a timestamp, sensor number, and physiological parameter value information; The data encryption module is configured to encrypt the pre-processed data using the AES algorithm, and the encrypted data is stored and transmitted in ciphertext form.

[0006] Furthermore, the intelligent control unit includes: a data fusion module configured to fuse the intragastric pressure, content properties, pH, and temperature data collected by the sensor array to eliminate redundancy and noise between the data; A model building module is configured to use a deep learning algorithm to build a gastrointestinal status model based on the fused multi-source data. The gastrointestinal status model includes gastric tone, peristalsis, and risk index. By training a large number of clinical gastrointestinal data samples, the gastrointestinal status model automatically learns the characteristic patterns of the gastrointestinal tract under different physiological and pathological states; an instruction generation module configured to use the data fused by the data fusion module as input features of the gastrointestinal status model, and dynamically generate personalized control instructions based on the severity of the patient's condition and gastrointestinal functional reserve assessed by the gastrointestinal status model, wherein the personalized control instructions include a negative pressure value adjustment range, a working mode, and a working duration; The closed-loop control module is configured to adopt an adaptive closed-loop control strategy according to the personalized control instructions to dynamically adjust the negative pressure value, working mode and working time of the intelligent gastrointestinal decompression device.

[0007] Furthermore, the intelligent protection unit includes: a risk prediction module configured to predict the risk of aspiration, mucosal damage or blockage in real time based on the data collected in real time by the data acquisition unit, and generate a corresponding risk signal if the risk of aspiration, mucosal damage or blockage exists; a mode switching module configured to receive the risk signal sent by the risk prediction module and automatically switch the intelligent gastrointestinal decompression device to a safe mode according to the risk signal; The risk warning module is configured to generate warning signals of different levels to alert medical staff according to the level of risk probability when the mode switching module switches to the safe mode.

[0008] Furthermore, based on the data collected in real time by the data acquisition unit, real-time prediction of aspiration, mucosal damage and blockage risks is performed, including: Retrieve data information collected by the sensor array in real time; Obtain parameter changes corresponding to each piece of data information based on the data information collected by the real-time sensor array; Obtaining a trend risk factor using the parameter change amount corresponding to each piece of data information, wherein the trend risk factor is used to characterize the risk contribution of the parameter change trend; Retrieve the current patient's physiological cycle phase angle; Retrieve the actual occurrence time period and theoretically predicted occurrence time point of each risk event of the patient's physiological events; Obtain the time deviation ΔT based on the actual occurrence time period of each risk event of the patient's physiological event and the theoretically predicted occurrence time point; The time deviation ΔT is combined with the current patient's physiological cycle phase angle to obtain a pattern risk factor for capturing the patient's physiological rhythm abnormalities; Furthermore, the trend risk factor and the pattern risk factor are used to obtain a comprehensive risk coefficient, including: Monitor the trend risk factors and pattern risk factors corresponding to patients in real time; Obtaining a trend risk factor change rate and a pattern risk factor change rate according to the trend risk factor and the pattern risk factor corresponding to the patient; Get the rate of change of negative pressure of the gastrointestinal decompression device; The trend risk factor change rate and the pattern risk factor change rate are combined with the negative pressure value change rate of the gastrointestinal decompression device to generate a comprehensive risk coefficient, and the larger the comprehensive risk coefficient is, the greater the risk is.

[0009] Furthermore, the closed-loop control module includes: The feedback adjustment module is configured to provide real-time feedback comparison between the actual operating status of the intelligent gastrointestinal decompression device and the preset personalized control instructions based on the real-time monitored gastric status data, and automatically adjust the parameters of the negative pressure value, working mode and working time according to the feedback results.

[0010] Furthermore, the mode switching module includes: a safety mode triggering module configured to receive a risk signal sent by the risk prediction module, including an aspiration risk signal, a mucosal injury risk signal, or a blockage risk signal; a parameter adjustment module configured to automatically adjust the operating parameters of the intelligent gastrointestinal decompression device according to the type of risk signal received, including reducing the negative pressure value, switching to an intermittent pulse mode, or pausing drainage; The state switching module is configured to switch the intelligent gastrointestinal decompression device from a normal working mode to a safe mode, and maintain the safe mode until the risk is eliminated.

[0011] Furthermore, the risk warning module includes: A multi-level warning module is configured to generate different levels of warning signals based on the risk probability, including low-risk warning, medium-risk warning, and high-risk warning, and alert medical staff through different warning methods, including but not limited to sound, vibration, and text prompts; A risk recording module is configured to record the results of each risk prediction, the time when the risk occurs, the risk type and level, and the safety measures taken; The risk analysis module is configured to perform statistical analysis on the risk data recorded by the risk recording module, generate a risk report, and transmit the risk report to the terminal device of the medical staff through wireless communication technology, so that the medical staff can understand the patient's risk trend and the operating status of the device.

[0012] A sensor-based intelligent gastrointestinal decompression control method, applicable to a sensor-based intelligent gastrointestinal decompression device, comprises the following steps: A sensor array is integrated into the gastric tube and calibrated regularly to monitor physiological parameters such as intragastric pressure, content properties, pH, and temperature in real time through the integrated sensor array; The intragastric pressure, content properties, pH and temperature data collected by the sensor array are fused and processed; Based on the fused multi-source data, a deep learning algorithm is used to build a gastrointestinal state model. By training a large number of clinical gastrointestinal data samples, the characteristic patterns of the gastrointestinal tract in different physiological and pathological states are automatically learned. The gastrointestinal status model is used to assess the severity of the patient's condition and gastrointestinal functional reserve, and dynamically generate personalized control instructions; Based on the real-time monitored gastric status data, the actual operating status of the intelligent gastrointestinal decompression device is compared with the preset personalized control instructions in real time, and the negative pressure value, working mode and working time parameters are automatically adjusted based on the feedback results; Based on real-time collected data, the risk of aspiration, mucosal damage or blockage is predicted in real time. If there is a risk, a corresponding risk signal is generated to alert medical staff.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The data acquisition unit of the present invention monitors multiple parameters in the stomach in real time through an integrated sensor array, and can ensure data accuracy by regularly calibrating the sensors. The intelligent control unit builds a gastrointestinal status model based on deep learning, accurately assesses the condition and generates personalized control instructions, and adopts an adaptive closed-loop control strategy to ensure the accuracy and stability of the operation of the intelligent gastrointestinal decompression device. The intelligent protection unit can predict risks in real time and automatically switch to a safe mode. At the same time, the multi-level warning module generates different levels of warning signals according to the risk probability, reminding medical staff to deal with it in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the structure of the sensor-based intelligent gastrointestinal decompression device of the present invention; Figure 2 This is a flow chart of the sensor-based intelligent gastrointestinal decompression control method of the present invention. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0016] In order to solve the technical problem that existing gastrointestinal decompression devices usually adopt fixed negative pressure values ​​and working modes, and cannot be adjusted according to the specific conditions of patients, thus it is difficult to meet the needs of different patients in different physiological and pathological conditions, please refer to Figure 1-Figure 2, this embodiment provides the following technical solutions: A sensor-based intelligent gastrointestinal decompression device, comprising a data acquisition unit, an intelligent control unit, and an intelligent protection unit; The data acquisition unit is configured to be integrated with a sensor array on the gastric tube, and the sensor array is used to monitor the intragastric pressure, content properties, pH and temperature in real time; Based on the data collected in real time by the data acquisition unit, the risk of aspiration, mucosal injury and blockage is predicted in real time, providing a more physiological and pathological risk assessment result for real-time clinical intervention. If there is a risk, a corresponding risk signal is generated to alert medical staff. The intelligent control unit is configured to receive data collected by the sensor array, fuse multi-source data to construct a gastrointestinal state model, dynamically generate personalized control instructions, and then use an adaptive closed-loop control strategy to dynamically adjust the negative pressure value, working mode and working time of the intelligent gastrointestinal decompression device; The intelligent protection unit is configured such that the intelligent gastrointestinal decompression device has a built-in active protection mechanism for predicting the risk of aspiration, mucosal damage or blockage based on real-time data, and automatically switching to a safe mode while triggering an early warning; The intelligent gastrointestinal decompression device also includes a user interface for medical staff or patients to view the data collected by the sensor array and the control instructions generated by the intelligent control unit in real time, and to manually adjust some control parameters.

[0017] The technical effects of the above technical solution are: through the sensor array of the data acquisition unit, key parameters such as gastric pressure, content properties, pH and temperature are monitored in real time, and the collected data are transmitted to the intelligent control unit. By real-time monitoring of multiple parameters in the stomach, the negative pressure value and other working parameters can be accurately adjusted. After receiving the data, the intelligent control unit integrates multi-source data to build a gastrointestinal status model, and accurately evaluates the patient's gastrointestinal functional status to dynamically generate personalized control instructions to meet the individual needs of different patients, thereby improving the treatment effect and comfort. In addition, through the adaptive closed-loop control strategy, the negative pressure value and working parameters of the decompression device are automatically adjusted. The operating mode and working hours can further adapt to the specific situation and needs of the patient. The intelligent protection unit analyzes sensor data in real time to predict risks such as aspiration, mucosal damage or blockage. Once a potential risk is detected, the intelligent gastrointestinal decompression device will automatically switch to the preset safety mode and trigger an early warning signal to remind medical staff, effectively preventing complications such as aspiration, mucosal damage and blockage, ensuring patient safety, and providing medical staff and patients with an intuitive user interface to view monitoring data and control instructions in real time, and allowing medical staff to manually adjust some control parameters according to clinical judgment, thereby increasing the flexibility and adaptability of the use of the intelligent gastrointestinal decompression device.

[0018] Data acquisition unit, including: a data calibration module configured to periodically calibrate the sensor array and perform a verification test after calibration to check whether the sensor output is within an allowable error range by repeatedly applying a known standard value. If the error range is exceeded, a recalibration prompt is issued and the number of calibration failures is recorded. If the number of recorded failures exceeds a preset number (e.g., 3 times), it is indicated that the sensor is faulty or damaged; The calibration process is as follows: the data calibration module sends a calibration instruction to the sensor array, and the sensor starts collecting data and sends it back to the data calibration module; The data calibration module compares the data collected by the sensor with the known standard value and analyzes the deviation between the data collected by the sensor and the known standard value; Based on the deviation analysis results, the data calibration module adjusts the calibration parameters of the sensor array, including correcting the sensitivity and offset parameters of the sensors; The data processing module is configured to pre-process the data collected by the sensor array, including: Through data cleaning, outliers, erroneous data and redundant data are removed from the sensor collected data; Adopting adaptive filtering algorithm to remove high-frequency noise and low-frequency drift in sensor collected data in real time; A microprocessor is used to perform preliminary operations on the data, converting the raw data into physiological parameter values, and encoding and packaging the data using a coding format to package the data into a data frame. Each data frame contains a timestamp, sensor number, and physiological parameter value information; The data encryption module is configured to encrypt the pre-processed data using the AES algorithm, and the encrypted data is stored and transmitted in ciphertext form.

[0019] The technical effects of the above technical solution are as follows: the data calibration module can ensure the accuracy of the sensor output data by regularly calibrating the sensor array and conducting verification tests, thereby providing a reliable data basis for subsequent analysis and control. The data processing module can effectively remove outliers, erroneous data, redundant data, noise and drift through data cleaning and filtering algorithms, thereby improving the accuracy and availability of the data. The micro-microprocessor then performs preliminary calculations and encoding and packaging on the data, converts the raw data into physiological parameter values, and forms a data frame containing timestamps, sensor numbers and physiological parameter value information, which is convenient for subsequent storage, transmission and analysis, thereby improving the efficiency of data processing. The data encryption module can effectively prevent data leakage and tampering by encrypting the pre-processed data using the AES algorithm, thereby protecting the patient's privacy and data security.

[0020] Intelligent control unit, including: a data fusion module configured to fuse the intragastric pressure, content properties, pH, and temperature data collected by the sensor array to eliminate redundancy and noise between the data; A model building module is configured to use a deep learning algorithm to build a gastrointestinal status model based on the fused multi-source data. The gastrointestinal status model includes gastric tone, peristalsis, and risk index. By training a large number of clinical gastrointestinal data samples, the gastrointestinal status model automatically learns the characteristic patterns of the gastrointestinal tract under different physiological and pathological states; an instruction generation module configured to use the data fused by the data fusion module as input features of the gastrointestinal status model, and dynamically generate personalized control instructions based on the severity of the patient's condition and gastrointestinal functional reserve assessed by the gastrointestinal status model, wherein the personalized control instructions include a negative pressure value adjustment range, a working mode, and a working duration; a closed-loop control module configured to dynamically adjust the negative pressure value, working mode, and working time of the intelligent gastrointestinal decompression device using an adaptive closed-loop control strategy according to the personalized control instructions; Among them, the closed-loop control module includes: The feedback adjustment module is configured to provide real-time feedback comparison between the actual operating status of the intelligent gastrointestinal decompression device and the preset personalized control instructions based on the real-time monitored gastric status data, and automatically adjust the parameters of the negative pressure value, working mode and working time according to the feedback results.

[0021] The technical effects of the above technical solution are as follows: the data fusion module can not only effectively eliminate the redundancy and noise between the data and make the data more accurate and complete by fusing the various data such as gastric pressure, content properties, pH and temperature collected by the sensor array, but also form a comprehensive description of the gastric status, avoiding the information loss or deviation that may be caused by a single data source, thereby improving the comprehensive understanding of the gastrointestinal status. The model construction module is based on a deep learning algorithm trained on a large number of clinical gastrointestinal data samples, which can automatically learn the characteristic patterns of the gastrointestinal tract under different physiological and pathological conditions, and construct a gastrointestinal status model including gastric tension, peristalsis and risk index. The constructed gastrointestinal status model can accurately evaluate the patient's gastrointestinal functional status, and through By learning the characteristic patterns of different physiological and pathological states, the gastrointestinal state model can adapt to various complex gastrointestinal states, thereby improving the universality and applicability of the intelligent gastrointestinal decompression device for different patients. The instruction generation module dynamically generates personalized control instructions based on the fused data and the evaluation results of the gastrointestinal state model. The generated personalized control instructions can meet the individualized needs of different patients and achieve precise treatment, thereby improving the treatment effect and patient comfort. The closed-loop control module adopts an adaptive closed-loop control strategy based on the personalized control instructions, and automatically adjusts the parameters through real-time feedback comparison of the actual operating status and the preset control instructions, so that the intelligent gastrointestinal decompression device can automatically optimize the operating parameters according to the patient's real-time status, thereby ensuring effective gastrointestinal decompression.

[0022] Intelligent protection unit, including: a risk prediction module configured to predict the risk of aspiration, mucosal damage or blockage in real time based on the data collected in real time by the data acquisition unit, and generate a corresponding risk signal if the risk of aspiration, mucosal damage or blockage exists; a mode switching module configured to receive the risk signal sent by the risk prediction module and automatically switch the intelligent gastrointestinal decompression device to a safe mode according to the risk signal; The risk warning module is configured to generate warning signals of different levels to alert medical staff according to the level of risk probability when the mode switching module switches to the safe mode.

[0023] The technical effect of the above technical solution is: the risk prediction module can predict the risk of aspiration, mucosal damage or blockage in real time by real-time analysis of data such as intragastric pressure, content properties, pH and temperature transmitted by the data acquisition unit. Once the risk is detected, a signal is generated to effectively prevent complications. After receiving the risk signal, the mode switching module immediately starts the safety mode, for example, reducing the negative pressure, switching to intermittent pulse mode or suspending drainage to ensure that the intelligent gastrointestinal decompression device can adjust the operating status in time when the risk occurs to avoid further harm to the patient. At the same time, the risk warning module will generate different levels of warning signals according to the risk probability to remind medical staff and ensure that medical staff can respond in time to reduce the adverse effects of the risk.

[0024] Specifically, based on the data collected in real time by the data collection unit, real-time prediction of the risk of aspiration, mucosal damage and blockage is performed, including: Retrieve data information collected by the sensor array in real time; Obtain parameter changes corresponding to each piece of data information based on the data information collected by the real-time sensor array; Obtaining a trend risk factor using the parameter change amount corresponding to each piece of data information, wherein the trend risk factor is used to characterize the risk contribution of the parameter change trend; The trend risk factor is obtained by the following formula: Where Г represents the trend risk factor; n represents the number of parameter types contained in the data information; ΔS i Indicates the parameter change corresponding to the i-th parameter; S i Indicates the standard change of the parameter corresponding to the i-th parameter; Retrieve the current patient's physiological cycle phase angle; Retrieve the actual occurrence time period and theoretically predicted occurrence time point of each risk event of the patient's physiological events; The time deviation ΔT is obtained based on the actual occurrence time period and theoretically predicted occurrence time point of each risk event (aspiration, mucosal injury or blockage) of the patient's physiological events; The time deviation ΔT is combined with the current patient's physiological cycle phase angle to obtain a pattern risk factor for capturing the patient's physiological rhythm abnormalities; The model risk factor is obtained by the following formula: Where U represents the pattern risk factor; θ represents the phase angle of the patient's current physiological cycle; m represents the total number of predicted risk events experienced by the patient; ΔT jT represents the time deviation between the actual occurrence time period of the physiological event of the j-th risk event and the theoretical predicted occurrence time point; ref，j Indicates the preset reference time window; The trend risk factor and the pattern risk factor are used to obtain a comprehensive risk coefficient.

[0025] The technical solution described above establishes a risk prediction system that integrates real-time parameter dynamic monitoring with circadian rhythm features. Trend risk factors precisely quantify the dynamic contribution of parameter change trends to risk. Pattern risk factors exploit the correlation between circadian cycle phase and the timing deviation of risk events, enabling refined and physiologically adaptive predictions of aspiration, mucosal injury, and obstruction risks. This provides more physiologically and pathologically informed risk assessments for real-time clinical intervention. Furthermore, the inverse tangent function applies nonlinear compression to the parameter change rate, avoiding the linear model's problem of ignoring small changes and over-amplifying large changes. For example, a linear model might miss a small abnormal parameter fluctuation (e.g., a physiological parameter slowly deviating from a threshold) because the change does not exceed the threshold. However, the arctan function captures this "trend anomaly," improving the accuracy of early risk identification. The pattern risk factor couples cos²(θ) (the circadian cycle phase angle) with the time deviation ΔT of risk events, converting circadian rhythm abnormalities (e.g., patients are more susceptible to aspiration during a specific circadian cycle phase) into quantifiable risk weights. Existing technologies often ignore the inherent relationship between "physiological rhythm and risk sequence". This model can accurately identify the changes in risk patterns caused by physiological cycle disorders, solve the problem of "risk differences under different physiological conditions due to abnormal parameters", and further improve the accuracy and efficiency of risk prediction. In addition, the trend risk factor is based on real-time sensor data (parameter change amount). ) Dynamic calculation, without waiting for full cycle data, can trigger risk trend judgment at the moment of parameter change, realizing low-latency mapping of "parameter-risk".

[0026] Furthermore, the trend risk factor formula uses the arctan function to nonlinearly map the ratio of the parameter change to the standard value, combined with the sign function sgn to distinguish the direction of change. Compared to the simple difference and ratio calculations used in the prior art, this method can more smoothly and accurately capture the risk impact of parameter change trends. This avoids oversensitivity to risk judgments when parameters fluctuate dramatically, while also effectively identifying potential risks accumulated from slowly changing trends. The pattern risk factor formula in the aforementioned technical solution incorporates the circadian phase angle θ, using cos²(θ) as the adjustment coefficient for the circadian rhythm state, and multiplies the ratio of the risk event time deviation to the reference window. While prior art approaches often fail to achieve such a deep quantitative integration of circadian rhythms and risk time series, the above-described technical solution in this embodiment can accurately identify changes in risk patterns caused by abnormal circadian phases, such as the frequent occurrence of risk events during specific circadian phases, making risk predictions more aligned with the characteristics of human circadian rhythms. The trend risk factor focuses on the dynamic trends of real-time parameters, while the pattern risk factor explores the association between circadian and time series anomalies. The two are not simply weighted, but are integrated based on independent mathematical models that characterize risk dimensions. Compared with the direct superposition of single-dimensional risk factors in the existing technology, this mechanism can reflect the complex causes of risks in a more comprehensive and hierarchical manner, thereby improving the accuracy of risk prediction and its clinical guidance value. Trend risk factors are continuously updated as real-time parameters change, and pattern risk factors are dynamically adjusted due to physiological cycle phases and risk event timing. The entire prediction system has real-time dynamic adaptability. Existing technologies often rely on static models or periodically updated models. The above technical solution in this embodiment can better track the dynamic changes in the patient's physiological state, timely capture the risk evolution trend, and provide a more timely and accurate basis for real-time intervention.

[0027] Specifically, the trend risk factor and the pattern risk factor are used to obtain a comprehensive risk coefficient, including: Monitor the trend risk factors and pattern risk factors corresponding to patients in real time; Obtaining a trend risk factor change rate and a pattern risk factor change rate according to the trend risk factor and the pattern risk factor corresponding to the patient; Get the rate of change of negative pressure of the gastrointestinal decompression device; The trend risk factor change rate and the pattern risk factor change rate are combined with the negative pressure value change rate of the gastrointestinal decompression device to generate a comprehensive risk coefficient, and the larger the comprehensive risk coefficient is, the greater the risk is.

[0028] The comprehensive risk coefficient is obtained by the following formula: Among them, Ф represents the comprehensive risk coefficient; t represents the current moment; U represents the pattern risk factor; Г represents the trend risk factor; P represents the rate of change of the negative pressure value; and Indicates the rate of change of trend risk factors and pattern risk factors; Indicates the rate of change of the negative pressure value of the gastrointestinal decompression device. Specifically, Construct the coupling relationship between trend and pattern risk change rate, and then combine it with the gastrointestinal decompression negative pressure change rate, and then combine it with Nonlinear correlation enables deep interaction of multi-dimensional risk influencing factors. Unlike the simple "factor superposition" of existing technologies, the technical solution described in this embodiment precisely quantifies the synergistic effects between factors—for example, how to amplify the impact of trend / pattern risks when negative pressure changes abnormally. This allows risk assessment to better align with the actual clinical scenarios in which multiple physiological / device parameters interact to influence risk.

[0029] The technical effect of the above technical solution is to build a comprehensive risk assessment system that integrates multi-dimensional dynamic factors (trend risk, pattern risk, and gastrointestinal decompression negative pressure). By accurately depicting the rate of change and interactive relationships of each factor, it achieves dynamic, precise, and clinically relevant quantitative predictions of the risks of aspiration, mucosal injury, and obstruction, providing medical staff with a real-time understanding of risk evolution and a precise basis for timely intervention. Through the "multi-factor change rate collaboration + device-physiological causal modeling", it solves the problems of misjudgment of single factor fluctuations and missed secondary device risks, and enhances accuracy. "Differential approximation + parallel computing" adapts to the real-time system computing power to improve prediction efficiency.

[0030] Furthermore, by introducing the trend risk factor change rate dΓ / dt and the pattern risk factor change rate dU / dt, this approach transcends the existing technology's reliance on static risk factor values ​​and instead focuses on the dynamic evolution of risk factors over time. This approach can capture the critical transition from "stable" to "sudden" risk factors, providing early warning of risk accumulation or outbreaks, rather than simply reacting passively to established risk states. The incorporation of the negative pressure change rate dP / dt of the gastrointestinal decompression device deeply integrates the dynamic parameters of clinical intervention devices with risk prediction. Existing technologies often view patient physiological parameters and device parameters in isolation. This technical solution in this embodiment embodies a "physiology-device linkage" risk logic: abnormal changes in device parameters (such as sudden changes in negative pressure) are directly incorporated into the risk factor calculation, assisting in identifying secondary risks caused by improper device intervention and improving the coverage of risk prediction across the entire clinical process. By leveraging nonlinear functions such as arctan (in trend / pattern risk factor calculations) and ln(1+P) (in comprehensive coefficient calculations), this approach provides flexible risk mapping for parameter changes. It avoids the excessive sensitivity or blunting of linear models in existing technologies to extreme values ​​and mutation values, can not only accurately identify the potential risks accumulated by small changes, but also reasonably constrain the impact of abnormal extreme values ​​on risk factors, making risk assessment more in line with the complex characteristics of "gradual change-mutation" of physiological signals.

[0031] Mode switching module, including: a safety mode triggering module configured to receive a risk signal sent by the risk prediction module, including an aspiration risk signal, a mucosal injury risk signal, or a blockage risk signal; a parameter adjustment module configured to automatically adjust the operating parameters of the intelligent gastrointestinal decompression device according to the type of risk signal received, including reducing the negative pressure value, switching to an intermittent pulse mode, or pausing drainage; The state switching module is configured to switch the intelligent gastrointestinal decompression device from a normal working mode to a safe mode, and maintain the safe mode until the risk is eliminated.

[0032] The technical effect of the above technical solution is: the safety mode trigger module can quickly respond to the signal sent by the risk prediction module by receiving the risk signal, and realize real-time risk management and control. The parameter adjustment module automatically adjusts the working parameters of the intelligent gastrointestinal decompression device according to different types of risk signals, for example, reducing the negative pressure value, switching to intermittent pulse mode or suspending drainage, thereby realizing precise risk management, thereby improving the adaptability and effectiveness of the intelligent gastrointestinal decompression device to different risk situations, and the state switching module switches the intelligent gastrointestinal decompression device to safety mode and maintains it until the risk is lifted, effectively ensuring that the intelligent gastrointestinal decompression device can operate stably when risks occur.

[0033] Risk early warning module, including: A multi-level warning module is configured to generate different levels of warning signals based on the risk probability, including low-risk warning, medium-risk warning, and high-risk warning, and alert medical staff through different warning methods, including but not limited to sound, vibration, and text prompts; A risk recording module is configured to record the results of each risk prediction, the time when the risk occurs, the risk type and level, and the safety measures taken; The risk analysis module is configured to perform statistical analysis on the risk data recorded by the risk recording module, generate a risk report, and transmit the risk report to the terminal device of the medical staff through wireless communication technology, so that the medical staff can understand the patient's risk trend and the operating status of the device.

[0034] The technical effects of the above technical scheme are as follows: the multi-level warning module can generate different levels of warning signals according to the level of risk probability, and remind medical staff through multiple methods such as sound, vibration and text prompts, so that medical staff can quickly pay attention to risks and take measures, thereby improving the emergency response speed. The risk recording module records the risk prediction results, occurrence time, type, level and safety measures taken in detail, providing a basis for medical accident tracing and risk assessment, and helping to analyze the causes of risks and optimize treatment plans. The risk analysis module can help medical staff accurately grasp the patient's risk trends and the operating status of the intelligent gastrointestinal decompression device by performing statistical analysis on the recorded risk data and generating risk reports, so as to adjust the treatment strategy in time. The risk warning module enhances the ability to manage patient risks through timely warning, detailed recording and accurate analysis of risks.

[0035] Specifically, this embodiment further proposes a sensor-based intelligent gastrointestinal decompression control method, which is applicable to a sensor-based intelligent gastrointestinal decompression device, and includes the following steps: A sensor array is integrated into the gastric tube and calibrated regularly to monitor physiological parameters such as intragastric pressure, content properties, pH, and temperature in real time through the integrated sensor array; The intragastric pressure, content properties, pH and temperature data collected by the sensor array are fused and processed; Based on the fused multi-source data, a deep learning algorithm is used to build a gastrointestinal state model. By training a large number of clinical gastrointestinal data samples, the characteristic patterns of the gastrointestinal tract in different physiological and pathological states are automatically learned. The gastrointestinal status model is used to assess the severity of the patient's condition and gastrointestinal functional reserve, and dynamically generate personalized control instructions; Based on the real-time monitored gastric status data, the actual operating status of the intelligent gastrointestinal decompression device is compared with the preset personalized control instructions in real time, and the negative pressure value, working mode and working time parameters are automatically adjusted based on the feedback results; Based on real-time collected data, the risk of aspiration, mucosal damage or blockage is predicted in real time. If there is a risk, a corresponding risk signal is generated to alert medical staff.

[0036] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0037] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A sensor-based intelligent gastrointestinal decompression device, characterized in that: The intelligent gastrointestinal decompression device includes a data acquisition unit, an intelligent control unit and an intelligent protection unit; The data acquisition unit is configured to be integrated with a sensor array on the gastric tube, and the sensor array is used to monitor the intragastric pressure, content properties, pH and temperature in real time; Based on the data collected in real time by the data acquisition unit, the risk of aspiration, mucosal injury and blockage is predicted in real time, providing a more physiological and pathological risk assessment result for real-time clinical intervention. If there is a risk, a corresponding risk signal is generated to alert medical staff. The intelligent control unit is configured to receive data collected by the sensor array, fuse multi-source data to construct a gastrointestinal state model, dynamically generate personalized control instructions, and then use an adaptive closed-loop control strategy to dynamically adjust the negative pressure value, working mode and working time of the intelligent gastrointestinal decompression device; The intelligent protection unit is configured as the intelligent gastrointestinal decompression device with a built-in active protection mechanism for predicting the risk of aspiration, mucosal damage or blockage based on real-time data, and automatically switching to a safe mode while triggering an early warning.

2. The sensor-based intelligent gastrointestinal decompression device according to claim 1, characterized in that: The data acquisition unit includes: a data calibration module configured to periodically calibrate the sensor array and perform a verification test after calibration to check whether the sensor output is within an allowable error range by repeatedly applying a known standard value. If the error range is exceeded, a reminder is given to recalibrate and the number of calibration failures is recorded. If the number of recorded failures exceeds a preset number, it is indicated that the sensor is faulty or damaged; The calibration process is as follows: the data calibration module sends a calibration instruction to the sensor array, and the sensor starts collecting data and sends it back to the data calibration module; The data calibration module compares the data collected by the sensor with the known standard value and analyzes the deviation between the data collected by the sensor and the known standard value; Based on the deviation analysis results, the data calibration module adjusts the calibration parameters of the sensor array, including correcting the sensitivity and offset parameters of the sensors; The data processing module is configured to pre-process the data collected by the sensor array, including: Through data cleaning, outliers, erroneous data and redundant data are removed from the sensor collected data; Adopting adaptive filtering algorithm to remove high-frequency noise and low-frequency drift in sensor collected data in real time; A microprocessor is used to perform preliminary operations on the data, converting the raw data into physiological parameter values, and encoding and packaging the data using a coding format to package the data into a data frame. Each data frame contains a timestamp, sensor number, and physiological parameter value information; The data encryption module is configured to encrypt the pre-processed data using the AES algorithm, and the encrypted data is stored and transmitted in ciphertext form.

3. The sensor-based intelligent gastrointestinal decompression device according to claim 1, characterized in that: The intelligent control unit comprises: a data fusion module configured to fuse the intragastric pressure, content properties, pH, and temperature data collected by the sensor array to eliminate redundancy and noise between the data; A model building module is configured to use a deep learning algorithm to build a gastrointestinal status model based on the fused multi-source data. The gastrointestinal status model includes gastric tone, peristalsis, and risk index. By training a large number of clinical gastrointestinal data samples, the gastrointestinal status model automatically learns the characteristic patterns of the gastrointestinal tract under different physiological and pathological states; an instruction generation module configured to use the data fused by the data fusion module as input features of the gastrointestinal status model, and dynamically generate personalized control instructions based on the severity of the patient's condition and gastrointestinal functional reserve assessed by the gastrointestinal status model, wherein the personalized control instructions include a negative pressure value adjustment range, a working mode, and a working duration; The closed-loop control module is configured to adopt an adaptive closed-loop control strategy according to the personalized control instructions to dynamically adjust the negative pressure value, working mode and working time of the intelligent gastrointestinal decompression device.

4. The sensor-based intelligent gastrointestinal decompression device according to claim 1, characterized in that: The intelligent protection unit comprises: a risk prediction module configured to predict the risk of aspiration, mucosal damage or blockage in real time based on the data collected in real time by the data acquisition unit, and generate a corresponding risk signal if the risk of aspiration, mucosal damage or blockage exists; a mode switching module configured to receive the risk signal sent by the risk prediction module and automatically switch the intelligent gastrointestinal decompression device to a safe mode according to the risk signal; The risk warning module is configured to generate warning signals of different levels to alert medical staff according to the level of risk probability when the mode switching module switches to the safe mode.

5. The sensor-based intelligent gastrointestinal decompression device according to claim 4, characterized in that: Based on the data collected in real time by the data acquisition unit, real-time prediction of aspiration, mucosal injury and blockage risks is performed, including: Retrieve data information collected by the sensor array in real time; Obtain parameter changes corresponding to each piece of data information based on the data information collected by the real-time sensor array; Obtaining a trend risk factor using the parameter change amount corresponding to each piece of data information, wherein the trend risk factor is used to characterize the risk contribution of the parameter change trend; Retrieve the current patient's physiological cycle phase angle; Retrieve the actual occurrence time period and theoretically predicted occurrence time point of each risk event of the patient's physiological events; Obtain the time deviation ΔT based on the actual occurrence time period of each risk event of the patient's physiological event and the theoretically predicted occurrence time point; The time deviation ΔT is combined with the current patient's physiological cycle phase angle to obtain a pattern risk factor for capturing the patient's physiological rhythm abnormalities; The trend risk factor and the pattern risk factor are used to obtain a comprehensive risk coefficient.

6. The sensor-based intelligent gastrointestinal decompression device according to claim 5, characterized in that: The trend risk factor and pattern risk factor are used to obtain a comprehensive risk factor, including: Monitor the trend risk factors and pattern risk factors corresponding to patients in real time; Obtaining a trend risk factor change rate and a pattern risk factor change rate according to the trend risk factor and the pattern risk factor corresponding to the patient; Get the rate of change of negative pressure of the gastrointestinal decompression device; The trend risk factor change rate and the pattern risk factor change rate are combined with the negative pressure value change rate of the gastrointestinal decompression device to generate a comprehensive risk coefficient, and the larger the comprehensive risk coefficient is, the greater the risk is.

7. The sensor-based intelligent gastrointestinal decompression device according to claim 3, characterized in that: The closed-loop control module includes: The feedback adjustment module is configured to provide real-time feedback and comparison between the actual operating status of the intelligent gastrointestinal decompression device and the preset personalized control instructions based on the real-time monitored gastric status data, and automatically adjust the negative pressure value, working mode and working time according to the feedback results.

8. The sensor-based intelligent gastrointestinal decompression device according to claim 4, characterized in that: The mode switching module includes: a safety mode triggering module configured to receive a risk signal sent by the risk prediction module, including an aspiration risk signal, a mucosal injury risk signal, or a blockage risk signal; a parameter adjustment module configured to automatically adjust the operating parameters of the intelligent gastrointestinal decompression device according to the type of risk signal received, including reducing the negative pressure value, switching to an intermittent pulse mode, or pausing drainage; The state switching module is configured to switch the intelligent gastrointestinal decompression device from a normal working mode to a safe mode, and maintain the safe mode until the risk is eliminated.

9. The sensor-based intelligent gastrointestinal decompression device according to claim 4, characterized in that: The risk warning module includes: A multi-level warning module is configured to generate different levels of warning signals based on the risk probability, including low-risk warning, medium-risk warning, and high-risk warning, and alert medical staff through different warning methods, including but not limited to sound, vibration, and text prompts; A risk recording module is configured to record the results of each risk prediction, the time when the risk occurs, the risk type and level, and the safety measures taken; The risk analysis module is configured to perform statistical analysis on the risk data recorded by the risk recording module, generate a risk report, and transmit the risk report to the terminal device of the medical staff through wireless communication technology, so that the medical staff can understand the patient's risk trend and the operating status of the device.

10. A sensor-based intelligent gastrointestinal decompression control method, applicable to a sensor-based intelligent gastrointestinal decompression device according to any one of claims 1 to 9, characterized in that: The following steps are involved: A sensor array is integrated into the gastric tube and calibrated regularly to monitor physiological parameters such as intragastric pressure, content properties, pH, and temperature in real time through the integrated sensor array; The intragastric pressure, content properties, pH and temperature data collected by the sensor array are fused and processed; Based on the fused multi-source data, a deep learning algorithm is used to build a gastrointestinal state model. By training a large number of clinical gastrointestinal data samples, the characteristic patterns of the gastrointestinal tract in different physiological and pathological states are automatically learned. The gastrointestinal status model is used to assess the severity of the patient's condition and gastrointestinal functional reserve, and dynamically generate personalized control instructions; Based on the real-time monitored gastric status data, the actual operating status of the intelligent gastrointestinal decompression device is compared with the preset personalized control instructions in real time, and the negative pressure value, working mode and working time parameters are automatically adjusted based on the feedback results; Based on real-time collected data, the risk of aspiration, mucosal damage or blockage is predicted in real time. If there is a risk, a corresponding risk signal is generated to alert medical staff.

Images