Intelligent anesthesia atomization equipment state monitoring method and system

Abstract

The invention relates to the technical field of medical equipment monitoring, in particular to an intelligent anesthesia atomization equipment state monitoring method and system. The method comprises the following steps: firstly, collecting atomization flow, particle size distribution, motor vibration frequency, liquid medicine temperature and pipeline pressure, and constructing an equipment state feature set; synchronously collecting electroencephalogram signals of a patient and constructing electroencephalogram state vectors; inputting the feature set into a support vector machine model to identify the running state of the equipment, and judging whether abnormity exists or not; when the equipment state is normal, an atomization efficiency attenuation equation is constructed based on continuous data, and the future efficiency trend is predicted; the prediction result and the electroencephalogram state are input into an anesthesia depth index dynamic balance model, and the adaptability of atomization output to the sedation reaction of the patient is evaluated; and when the efficiency attenuation rate is too high or the dynamic error exceeds a threshold value, generating a prompt signal and uploading the prompt signal to a remote platform. According to the invention, equipment state monitoring, efficiency prediction and neural feedback linkage control are realized, and intraoperative safety and postoperative evaluation precision are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment monitoring, and in particular to an intelligent anesthesia atomization equipment status monitoring method and system. Background Art

[0002] With the widespread application of anesthetic atomization technology in non-invasive sedation, preoperative induction, and local anesthesia, ensuring the stability and safety of the atomization process has become a key clinical concern. This invention aims to enhance the intelligent monitoring capabilities of device operation, enabling the simultaneous perception and dynamic regulation of anesthetic efficacy and the patient's physiological state.

[0003] Existing anesthesia nebulizer equipment generally relies on manual observation of atomized particles, device noise, or simple threshold alarms to determine its status. This makes it difficult to promptly identify critical issues such as particle anomalies, uneven atomization, and pipeline blockages, and it also lacks a predictive maintenance mechanism for device operating trends. Furthermore, while existing monitoring systems can collect physiological parameters such as heart rate and blood oxygen levels, they lack a mechanism for coupling these with device status, making it difficult to dynamically assess the compatibility of atomizer output with the patient's sedation response. This can easily lead to deviations in anesthesia depth control and pose potential risks.

[0004] To this end, the present invention proposes an intelligent anesthesia atomization equipment status monitoring method and system. Summary of the Invention

[0005] The present invention provides an intelligent anesthetic atomization equipment status monitoring method and system, aiming to enhance the intelligent monitoring capability of equipment operation and realize the synchronous perception and dynamic regulation of anesthetic efficacy and patient physiological status.

[0006] To achieve the above object, the present invention provides the following technical solutions: An intelligent anesthetic atomization equipment status monitoring method, comprising: Collecting the operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, to construct a device state feature set. Simultaneously collecting the patient's EEG signals, extracting spectral features reflecting the sedation state, and constructing an EEG state vector. Input the device status feature set into the status recognition model to identify the current operating status of the device. When it is identified as an abnormal state, a prompt signal is generated and the device status information is uploaded; When it is identified as a normal state, the atomization-related characteristic indicators are calculated based on the continuous operation data and substituted into the atomization efficiency attenuation equation to predict the future change trend of the atomization efficiency; Inputting the atomization efficiency and the EEG state vector into a dynamic balance model of anesthesia depth index to calculate the dynamic error between atomization output and neural response; When the performance decay rate exceeds the set threshold, or the dynamic error continues to exceed the target range, a prompt signal is generated, and the atomization prediction results, EEG state vector and error information are uploaded to the remote monitoring platform.

[0007] As a preferred technical solution of the invention, the step of constructing the device status feature set includes: Multiple sets of sensors installed inside the atomizer and at the gas path nodes collect atomization flow sensor signals, particle size detection signals, motor acceleration sensor signals, liquid medicine temperature signals, and gas path pressure signals respectively; Filter, de-noise, and standardize each signal to extract statistical characteristic parameters that characterize the equipment's operating status, including flow rate mean, particle size peak, vibration spectrum energy, temperature gradient, and pressure fluctuation amplitude; The above-mentioned characteristic parameters are combined into a multi-dimensional state vector, which is input into a subsequent state recognition model as the device state feature set.

[0008] As a preferred technical solution of the invention, the step of constructing the EEG state vector includes: Real-time EEG signals are collected through EEG electrodes; Filtering and removing artifacts from the EEG signal to extract power spectrum density values ​​of the α-wave and β-wave frequency bands; Calculating an EEG sedation index based on the ratio of alpha wave to beta wave power, wherein the sedation index is used to reflect the patient's anesthesia depth; The α wave power, β wave power and EEG sedation index are used to form an EEG state vector, which is used to evaluate the adaptability of atomization efficiency and patient neural response.

[0009] As a preferred technical solution of the invention, the state recognition model is a support vector machine classification model constructed based on a support vector set, and its specific implementation includes: Collect historical operating parameters of the equipment under various operating states, build a training sample set, standardize the equipment status feature set in the sample, and attach corresponding operating status labels; Selecting a support vector set for constructing a support boundary from a training sample set, and training the support vector machine classification model; The currently collected device status feature set is input into the support vector machine classification model, and the current operating status label of the device is output to determine whether there is uneven atomization, abnormal particles, depleted liquid, pipe blockage or dry burning.

[0010] As a preferred technical solution of the invention, the step of predicting the future change trend of atomization efficiency includes: The vibration frequency, gas line pressure and liquid temperature signals collected in a continuous time period are smoothed using the exponentially weighted moving average method to obtain the weighted average value of each indicator; Substituting the weighted average value into the atomization efficiency attenuation equation to calculate the atomization efficiency estimation values ​​corresponding to multiple time nodes; Constructing a trend curve of atomization efficiency changes in a future time period based on the estimated value to determine whether there is a risk of equipment efficiency decline; The calculation formula of the atomization efficiency attenuation equation is: ; in, is the initial effectiveness; They are the smoothed motor vibration frequency index, gas line pressure index and liquid medicine temperature index; are the attenuation coefficients of each parameter.

[0011] As a preferred technical solution of the invention, the steps for implementing the anesthesia depth index dynamic balance model include: Within the set time window, the predicted efficacy value at each moment in the atomization efficacy trend curve is paired with the EEG state vector of the corresponding period; Extracting EEG Calmness Index from EEG State Vector , and based on the current atomization efficiency value, target atomization efficiency, EEG sedation index and target sedation value, the following dynamic balance model is constructed: ; in, is the predicted value of efficacy; Target atomization efficiency; The preset target sedation value; and is the empirical adjustment coefficient, which is used to balance the relative influence of device output and neural response; By calculation The value of is obtained to obtain the dynamic error change rate between anesthetic efficacy and neural response, and to determine whether it is in the set adaptation range.

[0012] As a preferred technical solution of the invention, the steps of generating a prompt signal and uploading information include: Calculating the first-order difference of the predicted value of atomization efficacy within a continuous time period to obtain the efficacy decay rate and comparing it with a preset threshold; and determining whether the dynamic error change rate continuously exceeds the anesthesia depth adaptation interval; When any of the judgment conditions is met, an abnormality prompt signal is generated, and the prompt types include "performance attenuation abnormality", "adaptation imbalance" or "combined abnormality"; Construct a data package containing the current atomization performance trend prediction results, EEG state vector, dynamic error value and prompt type; The data packet is uploaded to the remote monitoring platform through the wireless communication module, and the timestamp and device ID information are attached.

[0013] The present invention also includes an intelligent anesthesia atomization equipment status monitoring system, comprising: A parameter acquisition module collects the operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, constructs a device state feature set, synchronously collects the patient's EEG signals, extracts spectral features reflecting the sedation state, and constructs an EEG state vector; A state recognition module is used to input the device state feature set into a state recognition model to identify the current operating state of the device. When an abnormal state is identified, a prompt signal is generated and the device state information is uploaded; The performance prediction module, when identified as a normal state, calculates atomization-related characteristic indicators based on continuous operation data, substitutes them into the atomization performance attenuation equation, and predicts the future trend of atomization performance; an error evaluation module, configured to input the atomization efficacy and the EEG state vector into a dynamic balance model of anesthesia depth index, and calculate a dynamic error between atomization output and neural response; The abnormal prompt and upload module generates a prompt signal when the performance decay rate exceeds the set threshold or the dynamic error continues to exceed the target range, and uploads the atomization prediction results, EEG state vector and error information to the remote monitoring platform to assist intraoperative judgment and postoperative evaluation.

[0014] The beneficial effects of the present invention are: 1. This invention constructs an atomization efficiency attenuation equation and combines it with multi-dimensional operating parameters to achieve dynamic modeling and trend prediction of anesthetic atomization performance. This approach is more continuous and forward-looking than existing solutions that rely on single parameter fluctuations. This mechanism not only improves the accuracy of performance evaluation but also provides data support for equipment maintenance and drug management. Combined with an abnormality warning mechanism, it significantly enhances the stability and control of the atomization process.

[0015] 2. This invention integrates and analyzes the patient's EEG state vector with the device's atomization performance trends and constructs a dynamic balance model for the depth of anesthesia index, enabling closed-loop monitoring of "device output-physiological response." This mechanism effectively overcomes the problem of existing devices focusing solely on physical parameters and ignoring individual patient responses, improving the precision and individualization of sedation control and possessing broad clinical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of an intelligent anesthesia atomization equipment status monitoring method of the present invention; Figure 2 The present invention is a structural diagram of an intelligent anesthesia atomization equipment status monitoring system. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0018] Example 1: Figure 1 The present invention provides an intelligent anesthesia atomization equipment status monitoring method, comprising: S100: collecting operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, constructing a device state feature set, synchronously collecting the patient's EEG signal, extracting spectral features reflecting the sedation state, and constructing an EEG state vector; Furthermore, the step of constructing the device status feature set includes: Multiple sets of sensors installed inside the atomizer and at the gas path nodes collect atomization flow sensor signals, particle size detection signals, motor acceleration sensor signals, liquid medicine temperature signals, and gas path pressure signals respectively; Filter, de-noise, and standardize each signal to extract statistical characteristic parameters that characterize the equipment's operating status, including flow rate mean, particle size peak, vibration spectrum energy, temperature gradient, and pressure fluctuation amplitude; The above-mentioned characteristic parameters are combined into a multi-dimensional state vector, which is input into a subsequent state recognition model as the device state feature set.

[0019] Specifically, the following five types of sensor components are deployed inside the anesthesia nebulizer equipment and at its key gas path nodes, which are used to collect the core parameters involved in the operation of the equipment: a thermal mass flow sensor is set at the spray outlet to obtain the instantaneous atomization flow value per second; a laser diffraction particle size detector is installed near the mist outlet to obtain the particle size distribution and the main peak position; a three-axis MEMS accelerometer is adhered to the atomizer motor housing to collect the motor vibration frequency and amplitude in real time; a thermocouple temperature sensor is embedded in the middle of the drug liquid circulation pipeline to record the temperature changes when the drug liquid passes through; a piezoresistive air pressure sensor is connected to the middle section of the connecting pipeline to collect real-time pressure data inside the pipeline.

[0020] After all collected signals are sent to the edge processing unit through the data acquisition module, the following operations are performed: bandpass filtering is performed on various sensor signals to filter out interference signals below 0.5Hz and above 50Hz; the sliding window method is applied to each parameter to calculate the average value and rate of change; and each parameter is normalized using the standard normalization method to facilitate subsequent unified modeling.

[0021] For each type of parameter, representative numerical features are extracted, including: the 5-second average of the atomization flow rate; the main peak diameter of the particle size; the main frequency of the motor vibration; the average heating rate of the liquid temperature; and the pipeline pressure fluctuation amplitude, that is, the difference between the maximum and minimum values.

[0022] The five characteristic parameters are arranged in a fixed order to form a multidimensional state vector of length 5. This vector is the device state feature set, which will be input into the state recognition model to determine whether the current device operating status is normal.

[0023] This feature construction method is portable and real-time, supporting operational status modeling for different types of anesthesia atomizer devices. By integrating multi-dimensional operational parameters to construct a state feature set, it provides a high-resolution input foundation for subsequent model recognition, enhancing the accuracy and generalization of device anomaly detection.

[0024] Furthermore, the step of constructing the EEG state vector includes: Real-time EEG signals are collected through EEG electrodes; Filtering and removing artifacts from the EEG signal to extract power spectrum density values ​​of the α-wave and β-wave frequency bands; Calculating an EEG sedation index based on the ratio of alpha wave to beta wave power, wherein the sedation index is used to reflect the patient's anesthesia depth; The α wave power, β wave power and EEG sedation index are used to form an EEG state vector, which is used to evaluate the adaptability of atomization efficiency and patient neural response.

[0025] Specifically, in this embodiment, a wireless multi-channel EEG acquisition device is used, with a 4-channel input, a sampling frequency set to 256 Hz, and a quantization accuracy of 16 bits. Disposable silver chloride gel electrodes are used as the electrode type, and the electrode impedance must be controlled within 5 kΩ to reduce noise. The electrode layout is in accordance with the international 10-20 system standard, and four channels, Fp1, Fp2, T3, and T4, are selected. The signal reference electrodes are placed on both earlobes, and the ground electrode is located at the midpoint of the forehead. The acquisition time window is set to slide and update every 10 seconds.

[0026] The collected raw EEG signals are preprocessed by the following steps: A third-order Butterworth bandpass filter was used to truncate the original signal in the 0.5–40 Hz frequency domain to remove power frequency interference and high-frequency myoelectric noise. Independent component analysis (ICA) was used to perform blind source separation on each EEG channel and extract independent components. The Pearson correlation coefficient between each component and the electrooculogram (EOG) channel was calculated to identify highly correlated artifacts (typically with correlation coefficients greater than 0.8 and peak voltages exceeding ±100 μV). These artifacts were then zeroed out and reconstructed to retain the valid brain-source components. The weighted average of the Fp1 and T3 signals was used as the left hemisphere signal channel, and the weighted average of the Fp2 and T4 signals was used as the right hemisphere signal channel to prevent local deviations from influencing subsequent features.

[0027] The purified EEG signal was processed using a fast Fourier transform (FFT) to extract the power spectral density of the following frequency bands: the α band and the β band. The analysis window was sliced ​​using a window function, and the power spectrum value for each frequency band was calculated using the integrated energy method in μV² / Hz.

[0028] The EEG calmness index is calculated by calculating the beta / alpha ratio (the ratio of beta wave power to alpha wave power): ; is the EEG sedation index obtained by spectrum analysis at the tth moment, The larger the value, the higher the alertness and the lighter the anesthesia; The smaller the value, the deeper the sedation state and the more adequate the anesthesia. Therefore, this ratio is used as the sedation index. Used for assessment of depth of anesthesia.

[0029] The above calculation results are spliced ​​into a three-dimensional vector to represent the EEG state at the current moment: ; This vector serves as an important input parameter for the system's subsequent physiological adaptability analysis and anesthetic effectiveness feedback.

[0030] By parameterizing EEG frequency band characteristics and constructing a standardized neural state expression, physiological feedback can form a comparable computational basis with device output, supporting individualized adaptation analysis.

[0031] S200: Inputting the device status feature set into a status recognition model to identify the current operating status of the device. When an abnormal state is identified, a prompt signal is generated and the device status information is uploaded; Furthermore, the state recognition model is a support vector machine classification model constructed based on a support vector set, and its specific implementation includes: Collect historical operating parameters of the equipment under various operating states, build a training sample set, standardize the equipment status feature set in the sample, and attach corresponding operating status labels; Selecting a support vector set for constructing a support boundary from a training sample set, and training the support vector machine classification model; The currently collected device status feature set is input into the support vector machine classification model, and the current operating status label of the device is output to determine whether there is uneven atomization, abnormal particles, depleted liquid, pipe blockage or dry burning.

[0032] Specifically, during the equipment testing phase, we manually set the equipment's operating conditions under different conditions to obtain multiple sets of data samples representing typical operating conditions. These conditions included normal atomization, uneven atomization, liquid depletion, pipe blockage, and dry-burn.

[0033] In each state, various sensors installed within the device acquire real-time operating parameters, including atomization flow rate, particle size, motor vibration frequency, liquid temperature, and pipeline pressure. This acquired parameter data is processed to extract statistical features that characterize state changes, including flow rate fluctuations, particle size stability, motor frequency fluctuations, temperature gradient changes, and pressure curve fluctuation trends.

[0034] The extracted feature parameters are then combined in a uniform order to form state feature vectors. Each set of feature vectors is labeled with its corresponding operating state to form a training sample set. To prevent the impact of dimensional differences between parameters on the model, the feature parameters in the sample are first standardized to ensure that they have similar data scales.

[0035] After constructing the training samples, the labeled status information in the training sample set is used to model the samples using the support vector classification principle. By selecting the most representative feature vectors in the samples as support boundary points, a support vector machine model for state classification is established. After model training is complete, the model is capable of determining the current device state based on the input feature vectors.

[0036] During the actual operation phase, when the device continues to work, the various state parameters of the current operation cycle are collected in real time through sensors, and the state features of the current cycle are extracted in accordance with the feature extraction method during training and combined into a new input vector.

[0037] This input vector is fed into a trained support vector machine classification model, which outputs a label for the device's current operating status based on the distribution of its eigenvalues. If the output label corresponds to an abnormal state, including normal atomization, uneven atomization, depleted liquid, blocked pipes, or dry burn, the system immediately generates a warning signal to notify the operator or activates the device's self-protection logic. This status information, along with the original parameters, is uploaded to a remote monitoring platform for intraoperative recording and postoperative analysis.

[0038] Support vector machines are used to classify and identify device status features, enabling intelligent judgment of multiple types of atomization anomalies. Compared with empirical methods, this method has stronger real-time performance and adaptability.

[0039] S300: When the normal state is identified, atomization-related characteristic indicators are calculated based on the continuous operation data, and the indicators are substituted into the atomization efficiency attenuation equation to predict the future change trend of atomization efficiency; Furthermore, the step of predicting the future change trend of atomization efficiency includes: The vibration frequency, gas line pressure and liquid temperature signals collected in a continuous time period are smoothed using the exponentially weighted moving average method to obtain the weighted average value of each indicator; Substituting the weighted average value into the atomization efficiency attenuation equation to calculate the atomization efficiency estimation values ​​corresponding to multiple time nodes; Constructing a trend curve of atomization efficiency changes in a future time period based on the estimated value to determine whether there is a risk of equipment efficiency decline; The calculation formula of the atomization efficiency attenuation equation is: ; in, is the initial effectiveness; They are the smoothed motor vibration frequency index, gas line pressure index and liquid medicine temperature index; are the attenuation coefficients of each parameter.

[0040] Specifically, during the continuous operation of the equipment, the system collects data including the motor vibration frequency according to the preset sampling period. , pipeline pressure and liquid temperature To enhance the robustness of the data, the system performs exponentially weighted moving average processing on the above raw time series data. The specific formula is as follows: ; in, Indicates the original sampling value of a certain operating parameter at the current moment; Represents the weighted average value at the current moment; For the smoothing factor, we collect historical atomization flow data under different patients and working conditions, set the candidate value range (such as 0.1 to 0.9), calculate the mean absolute error between each candidate value and the actual efficacy curve, and select the one with the smallest error. As the final use value.

[0041] Then, the system uses the three characteristic indicators obtained by the above smoothing process: vibration frequency ,pressure and temperature , substitute into the atomization efficiency attenuation equation to calculate the atomization efficiency value at the current moment , the specific formula is as follows: ; in, Indicates the theoretical maximum atomization efficiency under the initial optimal state of the equipment; They are used to characterize the influence of motor vibration frequency, liquid temperature and pipeline pressure on atomization efficiency respectively. The parameters are obtained by fitting historical operating data. Combined with multiple clinical sampling samples, the influence intensity of each operating parameter on atomization efficiency is determined through nonlinear regression analysis to ensure that the parameters are actually representative and repeatable.

[0042] The system continuously calculates multiple moments within the sliding time window. The system uses the value to construct a trend curve of atomization efficiency. A trend curve for atomization efficiency changes is constructed with value as the vertical axis and time as the horizontal axis, and a fitting analysis is performed on the slope changes of this curve at several future time points. When the trend curve shows a continuous downward trend within a set time period, or the estimated efficiency value corresponding to its endpoint falls below the preset operational safety threshold obtained from historical efficiency data under normal operating conditions, an early warning mechanism is triggered and the system is marked as in a state of potential efficiency degradation risk. Each estimated point on the trend curve is the predicted atomization efficiency value at the corresponding moment, which serves as the input basis for subsequent adaptive analysis.

[0043] By building a time-series continuous atomization efficiency trend model, the system can dynamically perceive the performance changes of the equipment, thereby shifting from passive detection to active prediction, significantly improving the forward-looking management and control capabilities of equipment operation stability.

[0044] S400: Inputting the atomization efficiency and the EEG state vector into a dynamic balance model of anesthesia depth index to calculate the dynamic error between atomization output and neural response; Furthermore, the steps for implementing the anesthesia depth index dynamic balance model include: Within the set time window, the predicted efficacy value at each moment in the atomization efficacy trend curve is paired with the EEG state vector of the corresponding period; Extracting EEG Calmness Index from EEG State Vector , and based on the current atomization efficiency value, target atomization efficiency, EEG sedation index and target sedation value, the following dynamic balance model is constructed: ; in, is the predicted value of efficacy; Target atomization efficiency; The preset target sedation value; and is the empirical adjustment coefficient, which is used to balance the relative influence of device output and neural response; By calculation The value of is obtained to obtain the dynamic error change rate between anesthetic efficacy and neural response, and to determine whether it is in the set adaptation range.

[0045] Specifically, within a set time window (e.g., 30 seconds), the system extracts the continuous atomization efficiency prediction value , which represents the predicted atomization efficiency value at the tth moment. Synchronously extract the corresponding EEG sedation index in the EEG state vector during this time period ,in is the EEG sedation index obtained by spectrum analysis at the t-th moment, obtained through the steps in S100.

[0046] Assume the target atomization efficiency is The target sedation value is , then the calculation formula of the dynamic error change rate is: ; in, It indicates the dynamic error change trend between anesthesia output and neural response, reflecting the immediate impact of enhanced or attenuated atomization efficiency on neural response. The empirical adjustment coefficient was determined by regression optimization method, including: selecting 50 cases of actual anesthesia nebulizer operation data and EEG sedation response data continuously recorded during surgery, covering two age groups of adults and elderly patients; constructing a sample set with nebulizer flow rate, particle mean, motor vibration and EEG sedation index, and minimizing the dynamic error sequence variance as the objective function, and fitting it using gradient descent method.

[0047] The system's continuous The calculation results are monitored. When the error change rate continues to exceed the set adaptation threshold range (for example, ±0.2) for more than a certain duration (for example, 20 seconds), the system determines that there is an adaptation abnormality in the current anesthesia state, indicating a possible risk of too shallow or too deep sedation.

[0048] Dynamic error analysis is performed by combining the changing relationship between atomization efficiency trends and EEG responses to achieve collaborative evaluation of human-machine status and improve the stability and individual adaptability of the anesthesia process.

[0049] S500: When the performance decay rate exceeds the set threshold, or the dynamic error continues to exceed the target range, a prompt signal is generated, and the atomization prediction results, EEG state vector and error information are uploaded to the remote monitoring platform to assist intraoperative judgment and postoperative evaluation.

[0050] Furthermore, the steps of generating a prompt signal and uploading information include: Calculating the first-order difference of the predicted value of atomization efficacy within a continuous time period to obtain the efficacy decay rate and comparing it with a preset threshold; and determining whether the dynamic error change rate continuously exceeds the anesthesia depth adaptation interval; When any of the judgment conditions is met, an abnormality prompt signal is generated, and the prompt types include "performance attenuation abnormality", "adaptation imbalance" or "combined abnormality"; Construct a data package containing the current atomization performance trend prediction results, EEG state vector, dynamic error value and prompt type; The data packet is uploaded to the remote monitoring platform through the wireless communication module, and the timestamp and device ID information are attached for intraoperative status warning and postoperative retrospective analysis.

[0051] Specifically, the system constructs a nebulizer efficacy attenuation trend curve using an exponentially weighted moving average method. The change between two adjacent predicted values ​​within each sampling period is calculated to obtain a first-order difference sequence, which characterizes the rate of change of nebulizer efficacy. The first-order difference mean within a fixed time window (e.g., 30 seconds) is set as the nebulizer efficacy attenuation rate and compared with a preset critical threshold (determined by device type, dosage form, and clinical data experience). When the attenuation rate exceeds this threshold, it indicates that the device is at significant risk of performance degradation.

[0052] At the same time, the system outputs the dynamic error based on the dynamic balance model. The system records the rate of change of the dynamic error for multiple consecutive time periods. If the error continuously deviates from the preset sedation adaptation interval, it is judged as abnormal neural adaptation.

[0053] Next, the system combines and analyzes the above two judgment processes and generates prompt information according to the following logic: if only the atomization efficiency attenuation rate exceeds the threshold, a "efficiency attenuation abnormality" prompt is generated; if only the dynamic error continuously exceeds the target range, a "adaptation imbalance" prompt is generated; if both are true at the same time, a "joint abnormality" prompt is generated.

[0054] To help doctors understand the device's operating status, intraoperative response risks, and postoperative analysis and review in real time, the system constructs the following data into a set of monitoring data packets: atomization efficiency trend curve within the current prediction period; EEG state vector during the same period; dynamic error value and its changing trend; abnormal type label and judgment result; current system timestamp and device identification information.

[0055] The data packet is uploaded to the remote monitoring platform. The platform receives uploaded information from each terminal device in real time and triggers a local alarm interface to prompt clinicians to respond promptly. Uploaded data is also archived for retrospective analysis of postoperative status, equipment maintenance records, and safety reviews of anesthesia procedures. By establishing a linkage mechanism between prompt generation and remote upload, the system achieves a synchronized closed loop between local judgment and remote response, ensuring timely communication of abnormal conditions and the formation of traceable records, enhancing the real-time and safety of clinical applications.

[0056] Example 2: In the thoracic surgery operating room of a certain hospital, a 65-year-old male patient underwent thoracoscopic lobectomy. The anesthesia was maintained using the ultrasonic atomization module integrated in the Dräger Perseus A500 anesthesia workstation and an intelligent anesthesia atomization equipment status monitoring system of the present invention. Figure 2 ,include: A parameter acquisition module collects the operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, constructs a device state feature set, synchronously collects the patient's EEG signals, extracts spectral features reflecting the sedation state, and constructs an EEG state vector; A state recognition module is used to input the device state feature set into a state recognition model to identify the current operating state of the device. When an abnormal state is identified, a prompt signal is generated and the device state information is uploaded; The performance prediction module, when identified as a normal state, calculates atomization-related characteristic indicators based on continuous operation data, substitutes them into the atomization performance attenuation equation, and predicts the future trend of atomization performance; an error evaluation module, configured to input the atomization efficacy and the EEG state vector into a dynamic balance model of anesthesia depth index, and calculate a dynamic error between atomization output and neural response; The abnormal prompt and upload module generates a prompt signal when the performance decay rate exceeds the set threshold or the dynamic error continues to exceed the target range, and uploads the atomization prediction results, EEG state vector and error information to the remote monitoring platform to assist intraoperative judgment and postoperative evaluation.

[0057] Specifically, 25 minutes after the start of the surgery, the following sensors of the parameter acquisition module collect data in real time: Atomization flow rate: A thermal mass flow meter is used to collect instantaneous flow rate. After Butterworth low-pass filtering, the mean and rate of change within a 5-second sliding window are calculated to obtain the current flow rate value of 0.82 L / min. Particle size distribution: The particle population was detected by laser diffractometer, and the main peak diameter was extracted to be 3.2 μm; Motor vibration: A three-axis accelerometer collects vibration signals and extracts the energy integral value of 0.15g² / Hz in the 20-200Hz frequency band through fast Fourier transform; Liquid temperature: The platinum resistance temperature sensor records the temperature gradient of the liquid as it flows through the pipeline, and the current value is 24.3°C; Pipeline pressure: Piezoresistive sensor monitors gas line pressure fluctuations, with a maximum fluctuation amplitude of 12mbar within a 10-second window.

[0058] The above data were processed using standard methods to obtain the atomization flow rate, vibration energy, temperature gradient, pipeline pressure, and peak particle size, respectively -0.25, 0.75, -1.0, 0.4, and 3.2. The five characteristic parameters were arranged in a fixed order to form a multidimensional state vector of length 5. This vector is the device state feature set: [-0.25, 3.2, 0.75, -1.0, 0.4].

[0059] The patient's EEG signals were acquired simultaneously using a four-channel EEG acquisition device (with Fp1, Fp2, T3, and T4 electrodes arranged according to a 10-20 system). After 0.5-40Hz bandpass filtering and independent component analysis to remove oculoculographic artifacts, the power spectral density values ​​for the alpha band and beta band were calculated to be 12.4μV² / Hz and 4.7μV² / Hz, respectively. The sedation index (B(t)) was then calculated to be 0.38. The time synchronization error between the device parameters and the EEG signals was controlled within ±50ms.

[0060] The above five-dimensional device status feature set was input into a pre-trained support vector machine classification model. This model was trained based on 200 sets of historical samples and uses a radial basis function kernel (penalty factor C = 1.0). The model output showed a 91% probability of normal status, a 5% probability of pipe blockage, and a 2% probability of dry-burning. The device was determined to be operating normally, and the performance prediction process began.

[0061] The vibration frequency, pipeline pressure and liquid temperature data within 30 consecutive seconds were smoothed using the exponential weighted moving average method (smoothing factor λ = 0.2) to obtain the weighted average values: vibration frequency index 0.73, pressure index 0.38, temperature index -0.98. Substitute the above indicators into the atomization efficiency attenuation equation: ; The initial efficiency =1.0, attenuation coefficient 、 、 (Determined through regression fitting of 100 sets of historical data).

[0062] Calculate the current performance value Based on this equation, it is further predicted that the efficacy will drop to 0.83 after 5 minutes, 0.80 after 10 minutes, and 0.77 after 15 minutes. EEG sedation index Enter the dynamic equilibrium model: ; The target atomization efficiency , target sedation index , adjustment coefficient 、 (Verified based on 50 clinical data).

[0063] Calculated dynamic error change rate Continuously monitor within three calculation cycles (15 seconds) All values ​​were less than -0.02, which was judged as "imbalance between anesthesia depth and device output adaptation."

[0064] The system triggers the following linkage responses: Local alarm: A red and yellow alarm icon will be displayed on the anesthesia workstation screen. The red area will be marked with "Atomization efficiency continues to decline (currently 0.85)" and the yellow area will be marked with "Anesthesia depth is too shallow (B(t)=0.38)"; Remote push: Send structured data packets to the remote monitoring platform through the communication module, including: Atomization efficiency prediction curve (0.85→0.83→0.80→0.77); EEG state vector [12.4, 4.7, 0.38]; Dynamic error value =-0.024; Device ID "Perseus-A500-034" and timestamp; Decision suggestion: The platform automatically generates an operation prompt: "Recommendation: Check the wear of the nebulizer component and increase the propofol infusion rate by 10%."

[0065] In this embodiment, the system uses dual-modal synchronous acquisition of device parameters and EEG signals. When the device's mechanical performance is normal, it predicts 15 minutes in advance that atomization efficiency will fall below the safety threshold (0.80), and simultaneously identifies the hidden risk of the patient's anesthesia depth being too shallow. After the clinical team adjusted the anesthesia plan accordingly, the patient's sedation index returned to the target range (0.35-0.40) within 8 minutes, effectively avoiding the risk of intraoperative awareness. Compared to traditional single-parameter monitoring methods, this solution shortens the abnormal response time to 8 seconds, reducing the number of intraoperative interventions.

[0066] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An intelligent anesthesia atomization equipment status monitoring method, characterized in that: include: Collecting the operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, to construct a device state feature set. Simultaneously collecting the patient's EEG signals, extracting spectral features reflecting the sedation state, and constructing an EEG state vector. Input the device status feature set into the status recognition model to identify the current operating status of the device. When it is identified as an abnormal state, a prompt signal is generated and the device status information is uploaded; When it is identified as a normal state, the atomization-related characteristic indicators are calculated based on the continuous operation data and substituted into the atomization efficiency attenuation equation to predict the future change trend of the atomization efficiency; Inputting the atomization efficiency and the EEG state vector into a dynamic balance model of anesthesia depth index to calculate the dynamic error between atomization output and neural response; When the performance decay rate exceeds the set threshold, or the dynamic error continues to exceed the target range, a prompt signal is generated, and the atomization prediction results, EEG state vector and error information are uploaded to the remote monitoring platform.

2. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The step of constructing the device status feature set includes: Multiple sets of sensors installed inside the atomizer and at the gas path nodes collect atomization flow sensor signals, particle size detection signals, motor acceleration sensor signals, liquid medicine temperature signals, and gas path pressure signals respectively; Filter, de-noise, and standardize each signal to extract statistical characteristic parameters that characterize the equipment's operating status, including flow rate mean, particle size peak, vibration spectrum energy, temperature gradient, and pressure fluctuation amplitude; The above-mentioned characteristic parameters are combined into a multi-dimensional state vector, which is input into a subsequent state recognition model as the device state feature set.

3. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The step of constructing the EEG state vector includes: Real-time EEG signals are collected through EEG electrodes; Filtering and removing artifacts from the EEG signal to extract power spectrum density values ​​of the α-wave and β-wave frequency bands; Calculating an EEG sedation index based on the ratio of alpha wave to beta wave power, wherein the sedation index is used to reflect the patient's anesthesia depth; The α wave power, β wave power and EEG sedation index are used to form an EEG state vector, which is used to evaluate the adaptability of atomization efficiency and patient neural response.

4. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The state recognition model is a support vector machine classification model built based on a support vector set, and its specific implementation includes: Collect historical operating parameters of the equipment under various operating states, build a training sample set, standardize the equipment status feature set in the sample, and attach corresponding operating status labels; Selecting a support vector set for constructing a support boundary from a training sample set, and training the support vector machine classification model; The currently collected device status feature set is input into the support vector machine classification model, and the current operating status label of the device is output to determine whether there is uneven atomization, abnormal particles, depleted liquid, pipe blockage or dry burning.

5. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The step of predicting the future change trend of atomization efficiency includes: The vibration frequency, gas line pressure and liquid temperature signals collected in a continuous time period are smoothed using the exponentially weighted moving average method to obtain the weighted average value of each indicator; Substituting the weighted average value into the atomization efficiency attenuation equation to calculate the atomization efficiency estimation values ​​corresponding to multiple time nodes; Constructing a trend curve of atomization efficiency changes in a future time period based on the estimated value to determine whether there is a risk of equipment efficiency decline; The calculation formula of the atomization efficiency attenuation equation is: ； in, is the initial effectiveness; They are the smoothed motor vibration frequency index, gas line pressure index and liquid medicine temperature index; are the attenuation coefficients of each parameter.

6. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The steps for implementing the anesthesia depth index dynamic balance model include: Within the set time window, the predicted efficacy value at each moment in the atomization efficacy trend curve is paired with the EEG state vector of the corresponding period; Extracting EEG Calmness Index from EEG State Vector , and based on the current atomization efficiency value, target atomization efficiency, EEG sedation index and target sedation value, the following dynamic balance model is constructed: ； in, is the predicted value of efficacy; Target atomization efficiency; The preset target sedation value; and is the empirical adjustment coefficient, which is used to balance the relative influence of device output and neural response; By calculation The value of is obtained to obtain the dynamic error change rate between anesthetic efficacy and neural response, and to determine whether it is in the set adaptation range.

7. The method for monitoring the state of an intelligent anesthesia atomization equipment according to claim 1, characterized in that: The steps of generating a prompt signal and uploading information include: Calculating the first-order difference of the predicted value of atomization efficacy within a continuous time period to obtain the efficacy decay rate and comparing it with a preset threshold; and determining whether the dynamic error change rate continuously exceeds the anesthesia depth adaptation interval; When any of the judgment conditions is met, an abnormality prompt signal is generated. The prompt types include "performance attenuation abnormality", "adaptation imbalance" or "combined abnormality"; Construct a data package containing the current atomization performance trend prediction results, EEG state vector, dynamic error value and prompt type; The data packet is uploaded to the remote monitoring platform through the wireless communication module, and the timestamp and device ID information are attached.

8. An intelligent anesthesia atomization equipment status monitoring system, characterized in that: For implementing the method steps according to any one of claims 1 to 7, the system comprises: A parameter acquisition module collects the operating parameters of the anesthesia atomizer device, including atomizer flow rate, particle size distribution, motor vibration frequency, drug solution temperature, and pipeline pressure, constructs a device state feature set, synchronously collects the patient's EEG signals, extracts spectral features reflecting the sedation state, and constructs an EEG state vector; A state recognition module is used to input the device state feature set into a state recognition model to identify the current operating state of the device. When an abnormal state is identified, a prompt signal is generated and the device state information is uploaded; The performance prediction module, when identified as a normal state, calculates atomization-related characteristic indicators based on continuous operation data, substitutes them into the atomization performance attenuation equation, and predicts the future trend of atomization performance; an error evaluation module, configured to input the atomization efficacy and the EEG state vector into a dynamic balance model of anesthesia depth index, and calculate a dynamic error between atomization output and neural response; The abnormal prompt and upload module generates a prompt signal when the performance decay rate exceeds the set threshold or the dynamic error continues to exceed the target range, and uploads the atomization prediction results, EEG state vector and error information to the remote monitoring platform to assist intraoperative judgment and postoperative evaluation.

Images