Artificial intelligence perioperative period anesthesia depth detection system

Abstract

The invention discloses an artificial intelligence perioperative period anesthesia depth detection system which comprises a data acquisition module which is provided with a multi-mode sensor and is used for acquiring physiological data of a patient in a perioperative period, and an artificial intelligence model module which is used for constructing an anesthesia depth prediction model based on a deep learning algorithm. The result output and display module is used for outputting the prediction result of the anesthesia depth in real time and visually displaying the prediction result on a display screen in the form of digits and charts, and the communication and interface module is provided with an interface for data communication with a hospital information system and an electronic medical record system. By means of multi-mode collection and innovative design, the anesthesia depth detection accuracy is improved, precise monitoring and intelligent auxiliary decision making are achieved, the method can be used for various scenes, the safety of patients in the perioperative period is effectively guaranteed, and the medical efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of anesthesia depth detection systems, and in particular to an artificial intelligence perioperative anesthesia depth detection system. Background Art

[0002] Accurate monitoring of anesthesia depth plays a vital role in perioperative medicine. During surgery, too shallow anesthesia may lead to increased awareness, pain perception, and stress response in patients, affecting the smooth progress of the operation and the patient's postoperative recovery; while too deep anesthesia may cause serious complications such as circulatory depression and respiratory depression, and even endanger the patient's life. Traditional anesthesia depth monitoring methods mainly rely on clinicians' observation and empirical judgment of patients' physiological signs such as heart rate, blood pressure, respiratory rate, etc. However, these indicators are affected by many factors and do not specifically reflect the depth of anesthesia. Their accuracy and reliability are limited.

[0003] In recent years, with the continuous advancement of science and technology, some new anesthesia depth monitoring technologies have gradually been applied in clinical practice. For example, methods based on electroencephalogram (EEG) analysis have received some attention. By analyzing the spectral characteristics and rhythm changes of EEG, the functional state of the brain and the depth of anesthesia can be reflected to a certain extent. However, single EEG signal monitoring also has limitations. It is easily affected by various interference factors, such as poor electrode contact and interference from electrosurgical equipment. In addition, there are individual differences in the EEG characteristics of different patients. It is difficult to achieve accurate anesthesia depth monitoring by relying solely on EEG analysis.

[0004] At the same time, existing anesthesia monitoring equipment is often relatively single in function and lacks the ability to comprehensively analyze multi-source physiological signals. During the perioperative period, the patient's physiological state is a complex system. In addition to EEG signals, physiological indicators such as electromyography (EMG), electrocardiogram (ECG), respiratory rate, body temperature, and pulse oxygen saturation are also closely related to the depth of anesthesia. Comprehensive consideration of these multimodal physiological signals is expected to improve the accuracy and reliability of anesthesia depth monitoring.

[0005] In addition, the current anesthesia monitoring system needs to be improved in terms of intelligence. The lack of real-time dynamic adjustment model makes it difficult to respond quickly and accurately according to the real-time changes of patients during surgery; in terms of early warning and decision support, most of them are simple threshold alarms, which cannot provide more targeted and practical suggestions for medical staff. Moreover, in the information age, the demand for interconnection between medical systems is increasing. The existing anesthesia monitoring equipment is still insufficient in integration and data sharing with hospital information systems (HIS), electronic medical record systems (EMR), etc., which limits the comprehensive utilization of medical data and the synergy of medical services.

[0006] Therefore, in order to overcome the shortcomings of existing anesthesia depth monitoring technology and equipment and improve the quality and safety of perioperative anesthesia management, it is urgently necessary to develop a new artificial intelligence perioperative anesthesia depth detection system that can integrate multi-source physiological signals, use advanced artificial intelligence algorithms for analysis and prediction, and have real-time dynamic adjustment, intelligent early warning and decision-making assistance, as well as good compatibility and remote monitoring functions to meet the needs of modern clinical anesthesia. Summary of the invention

[0007] The present invention proposes an artificial intelligence perioperative anesthesia depth detection system to solve the problems mentioned in the above-mentioned prior art.

[0008] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: an artificial intelligence perioperative anesthesia depth detection system, characterized in that it includes the following components:

[0009] Data acquisition module:

[0010] Equipped with multimodal sensors to collect the patient's perioperative physiological data, including but not limited to electroencephalogram (EEG), electromyography (EMG), electrocardiogram (ECG), respiratory rate, body temperature, and pulse oximetry;

[0011] The EEG sensor uses a high-density electrode array with no less than 32 electrodes to accurately collect subtle changes in EEG signals. The electrode material is silver / silver chloride material with good conductivity and biocompatibility.

[0012] The accuracy of other sensors should meet the corresponding medical industry standards, for example, the measurement accuracy of the pulse oximetry sensor should be within ±2%.

[0013] Signal preprocessing module:

[0014] Perform filtering, amplification, and denoising preprocessing operations on the collected physiological signals;

[0015] Adaptive filters are used to remove power frequency interference, electrooculographic artifacts and electromyographic artifacts in EEG signals to ensure signal purity;

[0016] Equipped with a signal amplifier, it can amplify weak physiological signals to a suitable amplitude range for subsequent processing, and the amplification factor can be adjusted within 10-1000 times.

[0017] Feature extraction module:

[0018] Use advanced signal processing algorithms to extract features related to anesthesia depth, such as spectral features, time domain features, and nonlinear features of EEG signals;

[0019] Calculate the power spectral density of the EEG signal, divide it into different frequency bands, such as delta wave (0.5-4Hz), theta wave (4-8Hz), alpha wave (8-13Hz), beta wave (13-30Hz), gamma wave (30Hz and above), and calculate the relative power of each frequency band;

[0020] The root mean square value and frequency characteristics of the EMG signal and the heart rate variability characteristics of the ECG signal are extracted as auxiliary features.

[0021] The following feature combination formula is introduced to comprehensively evaluate the anesthesia depth related features:

[0022] AI=w1×(P δ / P total )+w2×(P θ / P total )+w3×(RMS EMG )+w4×(HRV index )

[0023] Among them, AI represents the comprehensive anesthesia depth index, P δ , P θ are the power of delta wave and theta wave respectively, P total is the total power, RMS EMG is the root mean square value of the electromyographic signal, HRV index is the heart rate variability index, w1, w2, w3, and w4 are weight coefficients trained based on a large amount of clinical data. By adjusting these coefficients, the adaptability of the model to different patients and surgical conditions can be optimized.

[0024] Artificial intelligence model module:

[0025] Construct anesthesia depth prediction models based on deep learning algorithms, such as a combination of convolutional neural networks (CNN) and long short-term memory networks (LSTM);

[0026] The model was trained with a large amount of clinical data, covering patient data of different age groups, genders, surgical types, and types of anesthetic drugs. The ratio of training set, validation set, and test set was 7:2:1.

[0027] It has a model update mechanism that can update model parameters in real time or regularly based on new clinical data to adapt to changes in different clinical scenarios and patient groups.

[0028] Result output and display module:

[0029] The predicted results of anesthesia depth are output in real time and displayed intuitively on the display screen in the form of numbers and charts. The display interface should be concise and easy for medical staff to read;

[0030] When the anesthesia depth exceeds the preset safety range, an audible and visual alarm signal is issued, and the alarm threshold can be personalized according to the patient's specific situation and surgical stage;

[0031] It can store the patient's perioperative physiological data and anesthesia depth prediction results for no less than 1 year to facilitate subsequent data analysis and research.

[0032] Communication and interface modules:

[0033] It has an interface for data communication with the hospital information system (HIS) and the electronic medical record system (EMR), which can realize data transmission and sharing;

[0034] Use secure and reliable communication protocols, such as the HL7 standard protocol, to ensure the accuracy, integrity and confidentiality of data transmission and encrypt the transmitted data.

[0035] Preferably, the system also includes the following auxiliary modules:

[0036] Power management module: Provides stable power supply for all components of the system, with overvoltage protection, overcurrent protection and power failure backup power supply functions, ensuring that the system can still work normally for at least 30 minutes in emergency situations;

[0037] Equipment calibration module: Regularly calibrate the system's sensors and other equipment. The calibration procedures should comply with relevant national standards and industry specifications to ensure the accuracy and reliability of test data.

[0038] User authority management module: The system implements hierarchical authority management for system users. Users at different levels have different operating authorities. For example, doctors have the authority to view and modify parameters, while nurses only have the authority to view. This ensures the safety of system use and the confidentiality of data.

[0039] Preferably, the artificial intelligence perioperative anesthesia depth detection system has the following performance indicators:

[0040] Anesthesia depth prediction accuracy: The system's prediction accuracy of anesthesia depth is no less than 95% in different surgical types and patient groups;

[0041] Response time: The time interval from physiological signal acquisition to the output of anesthesia depth prediction results should not exceed 1 second to meet the requirements of real-time perioperative monitoring;

[0042] Stability: Within 48 hours of continuous operation, the fluctuation of various performance indicators should be within the allowable range, such as the acquisition error of EEG signals should not exceed ±5μV, and the consistency of anesthesia depth prediction results should not be less than 90%;

[0043] Compatibility: The system should be compatible with the hospital's existing various anesthesia equipment, monitoring equipment and information systems, seamlessly connect, and not affect the normal operation of other equipment.

[0044] Preferably, the system adopts the following innovative designs:

[0045] Multi-source data fusion and collaborative analysis: It not only integrates multiple physiological signals to detect the depth of anesthesia, but also combines the anesthetic drug dosage, infusion rate, and surgical operation step information during the operation for collaborative analysis to improve the accuracy and reliability of anesthesia depth prediction;

[0046] Real-time dynamic model adjustment: According to the patient's real-time physiological response and changes in anesthesia status during surgery, the parameters and structure of the artificial intelligence model are dynamically adjusted to achieve personalized and accurate anesthesia depth monitoring. For example, using online learning algorithms, the weights and biases of the model are updated in real time according to newly collected data, so that the model can quickly adapt to individual differences in patients and dynamic changes during surgery.

[0047] Intelligent early warning and decision support: In addition to the conventional anesthesia depth abnormal alarm, the system can also provide intelligent decision-making suggestions based on real-time monitoring data, such as recommending appropriate anesthetic drug adjustment plans, indicating possible anesthesia risks, and assisting medical staff in timely and effective anesthesia management. By establishing a decision tree model or a rule-based reasoning system, personalized decision-making suggestions are generated based on the patient's current anesthesia depth, physiological indicator change trends, and surgical progress.

[0048] Remote monitoring and consultation support: The system has a remote monitoring function. Doctors can view the patient's anesthesia status in real time on a remote terminal through the network, realize remote consultation and guidance, and improve the utilization efficiency of medical resources and the accessibility of medical services. The cloud computing-based architecture is used to achieve remote transmission and storage of data. Doctors can access the patient's anesthesia data and monitoring results anytime and anywhere through mobile phones and tablets.

[0049] Preferably, the application of an artificial intelligence perioperative anesthesia depth detection system in various types of surgeries can be widely applied to different types of surgical scenarios such as general surgery, cardiothoracic surgery, neurosurgery, orthopedic surgery, and obstetrics and gynecology surgery, providing comprehensive and accurate anesthesia depth detection services for perioperative anesthesia management.

[0050] Preferably, an artificial intelligence perioperative anesthesia depth detection system is used in various types of surgeries. The system can also be used in anesthesia recovery rooms and intensive care units (ICU) postoperative monitoring scenarios to continuously monitor and evaluate the patient's anesthesia awakening process and postoperative recovery status, and promptly detect and deal with possible anesthesia-related complications.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention ensures the comprehensiveness and accuracy of data collection by using high-density electrode array EEG sensors and a variety of high-precision physiological signal sensors, and can more accurately capture subtle changes in the patient's perioperative physiological state. The signal preprocessing module effectively removes interfering signals and provides pure data for subsequent analysis. The innovative feature combination formula and deep learning model greatly improve the accuracy of anesthesia depth prediction, which can reach more than 95%, providing strong protection for surgical safety.

[0053] The real-time dynamic adjustment model function can quickly adapt to the patient's real-time physiological response and surgical changes to achieve personalized and accurate monitoring. Intelligent early warning and decision-making assistance can not only detect abnormalities in a timely manner, but also provide targeted anesthetic drug adjustment suggestions, etc., effectively assisting medical staff in making decisions and improving the quality of anesthesia management. Remote monitoring and consultation support functions use cloud computing to achieve remote data transmission and storage, making it convenient for doctors to check patients' anesthesia conditions anytime and anywhere, improving the efficiency of medical resource utilization and service accessibility.

[0054] Multi-source data fusion and collaborative analysis comprehensively consider multiple physiological signals and surgery-related information, which comprehensively improves the accuracy and reliability of monitoring. The system has good compatibility and can seamlessly connect to the hospital's existing equipment and information systems to promote the comprehensive utilization and collaborative work of medical data. Overall, the present invention helps to improve the level of perioperative anesthesia management, reduce anesthesia risks, ensure patient safety, and promote the intelligent development of anesthesia monitoring technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a schematic block diagram of an artificial intelligence perioperative anesthesia depth detection system proposed by the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise" and "counterclockwise" indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0058] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0059] Reference Figure 1 :An artificial intelligence perioperative anesthesia depth detection system, characterized by comprising the following components:

[0060] Data acquisition module:

[0061] Equipped with multimodal sensors to collect the patient's perioperative physiological data, including but not limited to electroencephalogram (EEG), electromyography (EMG), electrocardiogram (ECG), respiratory rate, body temperature, and pulse oximetry;

[0062] The EEG sensor uses a high-density electrode array with no less than 32 electrodes to accurately collect subtle changes in EEG signals. The electrode material is silver / silver chloride material with good conductivity and biocompatibility.

[0063] The accuracy of other sensors should meet the corresponding medical industry standards, for example, the measurement accuracy of the pulse oximetry sensor should be within ±2%.

[0064] Signal preprocessing module:

[0065] Perform filtering, amplification, and denoising preprocessing operations on the collected physiological signals;

[0066] Adaptive filters are used to remove power frequency interference, electrooculographic artifacts and electromyographic artifacts in EEG signals to ensure signal purity;

[0067] Equipped with a signal amplifier, it can amplify weak physiological signals to a suitable amplitude range for subsequent processing, and the amplification factor can be adjusted within 10-1000 times.

[0068] Feature extraction module:

[0069] Use advanced signal processing algorithms to extract features related to anesthesia depth, such as spectral features, time domain features, and nonlinear features of EEG signals;

[0070] Calculate the power spectral density of the EEG signal, divide it into different frequency bands, such as delta wave (0.5-4Hz), theta wave (4-8Hz), alpha wave (8-13Hz), beta wave (13-30Hz), gamma wave (30Hz and above), and calculate the relative power of each frequency band;

[0071] The root mean square value and frequency characteristics of the EMG signal and the heart rate variability characteristics of the ECG signal are extracted as auxiliary features.

[0072] The following feature combination formula is introduced to comprehensively evaluate the anesthesia depth related features:

[0073] AI=w1×(P δ / P total )+w2×(P θ / P total )+w3×(RMS EMG )+w4×(HRV index )

[0074] Among them, AI represents the comprehensive anesthesia depth index, P δ , P θ are the power of delta wave and theta wave respectively, P total is the total power, RMS EMG is the root mean square value of the electromyographic signal, HRV index is the heart rate variability index, w1, w2, w3, and w4 are weight coefficients trained based on a large amount of clinical data. By adjusting these coefficients, the adaptability of the model to different patients and surgical conditions can be optimized.

[0075] Artificial intelligence model module:

[0076] Construct anesthesia depth prediction models based on deep learning algorithms, such as a combination of convolutional neural networks (CNN) and long short-term memory networks (LSTM);

[0077] The model was trained with a large amount of clinical data, covering patient data of different age groups, genders, surgical types, and types of anesthetic drugs. The ratio of training set, validation set, and test set was 7:2:1.

[0078] It has a model update mechanism that can update model parameters in real time or regularly based on new clinical data to adapt to changes in different clinical scenarios and patient groups.

[0079] Result output and display module:

[0080] The predicted results of anesthesia depth are output in real time and displayed intuitively on the display screen in the form of numbers and charts. The display interface should be concise and easy for medical staff to read;

[0081] When the anesthesia depth exceeds the preset safety range, an audible and visual alarm signal is issued, and the alarm threshold can be personalized according to the patient's specific situation and surgical stage;

[0082] It can store the patient's perioperative physiological data and anesthesia depth prediction results for no less than 1 year to facilitate subsequent data analysis and research.

[0083] Communication and interface modules:

[0084] It has an interface for data communication with the hospital information system (HIS) and the electronic medical record system (EMR), which can realize data transmission and sharing;

[0085] Use secure and reliable communication protocols, such as the HL7 standard protocol, to ensure the accuracy, integrity and confidentiality of data transmission and encrypt the transmitted data.

[0086] In the present invention, the system also includes the following auxiliary modules:

[0087] Power management module: Provides stable power supply for all components of the system, with overvoltage protection, overcurrent protection and power failure backup power supply functions, ensuring that the system can still work normally for at least 30 minutes in emergency situations;

[0088] Equipment calibration module: Regularly calibrate the system's sensors and other equipment. The calibration procedures should comply with relevant national standards and industry specifications to ensure the accuracy and reliability of test data.

[0089] User authority management module: The system implements hierarchical authority management for system users. Users at different levels have different operating authorities. For example, doctors have the authority to view and modify parameters, while nurses only have the authority to view. This ensures the safety of system use and the confidentiality of data.

[0090] In the present invention, the artificial intelligence perioperative anesthesia depth detection system has the following performance indicators:

[0091] Anesthesia depth prediction accuracy: The system's prediction accuracy of anesthesia depth is no less than 95% in different surgical types and patient groups;

[0092] Response time: The time interval from physiological signal acquisition to the output of anesthesia depth prediction results should not exceed 1 second to meet the requirements of real-time perioperative monitoring;

[0093] Stability: Within 48 hours of continuous operation, the fluctuation of various performance indicators should be within the allowable range, such as the acquisition error of EEG signals should not exceed ±5μV, and the consistency of anesthesia depth prediction results should not be less than 90%;

[0094] Compatibility: The system should be compatible with the hospital's existing various anesthesia equipment, monitoring equipment and information systems, seamlessly connect, and not affect the normal operation of other equipment.

[0095] In the present invention, the system adopts the following innovative designs:

[0096] Multi-source data fusion and collaborative analysis: It not only integrates multiple physiological signals to detect the depth of anesthesia, but also combines the anesthetic drug dosage, infusion rate, and surgical operation step information during the operation for collaborative analysis to improve the accuracy and reliability of anesthesia depth prediction;

[0097] Real-time dynamic model adjustment: According to the patient's real-time physiological response and changes in anesthesia status during surgery, the parameters and structure of the artificial intelligence model are dynamically adjusted to achieve personalized and accurate anesthesia depth monitoring. For example, using online learning algorithms, the weights and biases of the model are updated in real time according to newly collected data, so that the model can quickly adapt to individual differences in patients and dynamic changes during surgery.

[0098] Intelligent early warning and decision support: In addition to the conventional anesthesia depth abnormal alarm, the system can also provide intelligent decision-making suggestions based on real-time monitoring data, such as recommending appropriate anesthetic drug adjustment plans, indicating possible anesthesia risks, and assisting medical staff in timely and effective anesthesia management. By establishing a decision tree model or a rule-based reasoning system, personalized decision-making suggestions are generated based on the patient's current anesthesia depth, physiological indicator change trends, and surgical progress.

[0099] Remote monitoring and consultation support: The system has a remote monitoring function. Doctors can view the patient's anesthesia status in real time on a remote terminal through the network, realize remote consultation and guidance, and improve the utilization efficiency of medical resources and the accessibility of medical services. The cloud computing-based architecture is used to achieve remote transmission and storage of data. Doctors can access the patient's anesthesia data and monitoring results anytime and anywhere through mobile phones and tablets.

[0100] In the present invention, an artificial intelligence perioperative anesthesia depth detection system is applied in various types of surgeries, which can be widely applied to different types of surgical scenarios such as general surgery, cardiothoracic surgery, neurosurgery, orthopedic surgery, and obstetrics and gynecology surgery, providing comprehensive and accurate anesthesia depth detection services for perioperative anesthesia management.

[0101] In the present invention, an artificial intelligence perioperative anesthesia depth detection system is used in various types of surgeries. The system can also be used in anesthesia recovery rooms and intensive care units (ICU) postoperative monitoring scenarios to continuously monitor and evaluate the patient's anesthesia awakening process and postoperative recovery status, and promptly detect and deal with possible anesthesia-related complications.

[0102] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can replace or change the technical solution and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An artificial intelligence perioperative anesthesia depth detection system, characterized in that: It includes the following components: Data acquisition module: equipped with multimodal sensors to collect patients' perioperative physiological data, including EEG signals, EMG signals, ECG signals, respiratory rate, body temperature, and pulse oxygen saturation; The EEG sensor uses a high-density electrode array with no less than 32 electrodes to collect changes in EEG signals. The electrode material is silver or silver chloride material that is conductive and biocompatible; Signal preprocessing module: performs filtering, amplification, and denoising preprocessing operations on the collected physiological signals; Adaptive filter is used to remove power frequency interference, electrooculographic artifacts and electromyographic artifacts from EEG signals; Feature extraction module: uses signal processing algorithms to extract features related to anesthesia depth, including spectral features, time domain features, and nonlinear features of EEG signals; Calculate the power spectral density of the EEG signal, divide it into different frequency bands, and calculate the relative power of each frequency band; Extract the root mean square value and frequency characteristics of the EMG signal and the heart rate variability characteristics of the ECG signal as auxiliary features; Introduction The following feature combination formula is used to comprehensively evaluate the anesthesia depth related features: AI=w1×(P δ / P total )+w2×(P θ / P total )+w3×(RMS EMG )+w4×(HRV index ) Among them, AI represents the comprehensive anesthesia depth index, P δ , P θ are the power of delta wave and theta wave respectively, P total is the total power, RMS EMG is the root mean square value of the electromyographic signal, HRV index is the heart rate variability index, w1, w2, w3, and w4 are weight coefficients obtained through training based on a large amount of clinical data. By adjusting these coefficients, the adaptability of the model to different patients and surgical conditions can be optimized. Artificial intelligence model module: build an anesthesia depth prediction model based on deep learning algorithm, including a combined model of convolutional neural network and long short-term memory network; The model was trained with clinical data, including patient data of different age groups, genders, surgery types, and anesthetic drugs. The ratio of training set, validation set, and test set was 7:2:

1. Result output and display module: output the prediction results of anesthesia depth in real time and display them intuitively on the display screen in the form of numbers and charts; When the anesthesia depth exceeds the preset safety range, an audible and visual alarm signal is issued, and the alarm threshold can be set according to the patient's specific situation and the stage of surgery; Communication and interface module: It has an interface for data communication with the hospital information system and electronic medical record system to realize data transmission and sharing; Use secure and reliable communication protocols, including the HL7 standard protocol, to ensure the accuracy, integrity and confidentiality of data transmission and encrypt the transmitted data.

2. The artificial intelligence perioperative anesthesia depth detection system according to claim 1 is characterized in that: The system also includes the following auxiliary modules: Power management module: Provides stable power supply for all components of the system, with overvoltage protection, overcurrent protection and power failure backup power supply functions, ensuring that the system can still work normally for at least 30 minutes in emergency situations; Equipment calibration module: regularly calibrate the system's sensors and other equipment; User authority management module: The system implements hierarchical authority management for system users. Users at different levels have different operating authorities. Doctors have the authority to view and modify parameters, while nurses only have the authority to view.

3. An artificial intelligence perioperative anesthesia depth detection system according to any one of claims 1-2, characterized in that: It has the following performance indicators: Anesthesia depth prediction accuracy: The system's prediction accuracy of anesthesia depth is no less than 95% in different surgical types and patient groups; Response time: The time interval from physiological signal acquisition to the output of anesthesia depth prediction results should not exceed 1 second to meet the requirements of real-time perioperative monitoring; Stability: Within 48 hours of continuous operation, the fluctuation of various performance indicators should be within the allowable range, the acquisition error of EEG signals should not exceed ±5μV, and the consistency of anesthesia depth prediction results should not be less than 90%; Compatibility: The system should be compatible with the hospital's existing various anesthesia equipment, monitoring equipment and information systems without affecting the normal operation of other equipment.

4. An artificial intelligence perioperative anesthesia depth detection system according to any one of claims 1 to 3, characterized in that: Also includes: Real-time dynamic model adjustment: According to the patient's real-time physiological response and changes in anesthesia status during surgery, the parameters and structure of the artificial intelligence model are adjusted to achieve personalized and accurate anesthesia depth monitoring. Specifically, the online learning algorithm is used to update the model's weights and biases in real time according to the newly collected data, so that the model can quickly adapt to the individual differences of patients and dynamic changes during surgery; Intelligent early warning and decision-making assistance module: Provide intelligent decision-making suggestions based on real-time monitoring data, including recommending appropriate anesthetic drug adjustment plans, indicating possible anesthesia risks, and assisting medical staff in timely and effective anesthesia management; Remote monitoring and consultation support module: Doctors can view the patient's anesthesia status in real time on the remote terminal through the network, realize remote consultation and guidance, and adopt a cloud computing-based architecture to achieve remote transmission and storage of data. Doctors can access the patient's anesthesia data and monitoring results through mobile phones and tablets.

5. An application of the artificial intelligence perioperative anesthesia depth detection system according to any one of claims 1 to 4 in various surgeries, characterized in that: Used in different types of surgical scenarios including general surgery, cardiothoracic surgery, neurosurgery, orthopedic surgery, and obstetrics and gynecology.

6. The application of the artificial intelligence perioperative anesthesia depth detection system according to claim 5 is characterized in that: It is also used in postoperative monitoring scenarios in anesthesia recovery rooms and intensive care units to continuously monitor and evaluate the patient's anesthesia awakening process and postoperative recovery status.

Images