Anesthesia state evaluation method and system based on electroencephalogram data analysis

Abstract

The invention relates to the technical field of anesthesia state evaluation, and particularly discloses an anesthesia state evaluation method and system based on electroencephalogram data analysis. Comprising an electroencephalogram data acquisition module for collecting real-time electroencephalogram data including voltage, frequency and waveform characteristics of electroencephalogram signals, a risk assessment module, an early warning and control module for identifying a potential anesthesia risk state from an anesthesia state assessment scheme, and a risk early warning information sending module for sending risk early warning information to an anesthetist according to an anesthesia risk assessment result. According to the method, the voltage, frequency and waveform characteristic parameters of the electroencephalogram signals are collected in real time, the electroencephalogram signal characteristic indexes and the real-time electroencephalogram state diagram are generated, on the basis, the anesthesia state is analyzed based on the characteristic indexes and characteristic grades of the electroencephalogram signals, the anesthesia depth limited state and the potential risk state are recognized, and the accuracy of anesthesia is improved. And an anesthesia state assessment scheme based on the electroencephalogram characteristic mode association network is established, so that the anesthesia depth and the high-risk state are accurately assessed, and the anesthesia state assessment capability is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of anesthesia state assessment, and in particular, relates to an anesthesia state assessment method and system based on electroencephalogram (EEG) data analysis. Background Art

[0002] The field of anesthesia state assessment technology primarily involves key technologies such as EEG data collection, feature extraction, anesthesia depth analysis, and state assessment. By acquiring the patient's EEG signals through devices such as EEG electrodes and analyzing EEG characteristics using signal processing and artificial intelligence methods, the patient's anesthesia depth can be accurately assessed and monitored.

[0003] However, existing technologies have difficulty effectively distinguishing different stages of anesthesia depth in practice, making it difficult to accurately assess a patient's anesthesia status. This can lead to over- or under-anesthesia, compromising surgical safety and patient recovery, and increasing surgical risks and complications. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of the present invention is to provide an anesthetic state assessment method and system based on EEG data analysis, which solves the problems existing in the prior art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] Anesthesia status assessment system based on EEG data analysis, including:

[0007] The EEG data acquisition module collects real-time EEG data, including the voltage, frequency, and waveform characteristics of the EEG signal, calculates the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generates a real-time state diagram of the EEG signal; based on the real-time EEG state diagram, analyzes the characteristic indicators of the EEG signal and generates EEG characteristic analysis results;

[0008] A real-time analysis module determines the characteristic pattern of the EEG signal based on the EEG characteristic analysis results and obtains a preliminary evaluation result of the anesthetic state; based on the preliminary evaluation result, adjusts the evaluation model weight, repeatedly iterates to accurately evaluate the anesthetic state, and generates an anesthetic state evaluation plan;

[0009] a risk assessment module that identifies potential anesthesia risk states from the anesthesia state assessment scheme, analyzes the risks, detects abnormal EEG patterns, and obtains potential risk state analysis results; based on the potential risk state analysis results, assesses the probability of anesthesia risk occurrence and the estimated impact, and generates an anesthesia risk assessment result;

[0010] The early warning and control module sends risk early warning information to the anesthesiologist based on the anesthesia risk assessment results to obtain risk early warning and control outputs.

[0011] Preferably, the acquiring step of the real-time brain electrical signal state diagram is:

[0012] The voltage, frequency and waveform characteristics of the real-time brain electrical signal of the patient are collected, the waveform classification coefficient is determined according to the waveform characteristics, and the brain electrical signal basic data is formed;

[0013] According to the brain electrical signal basic data, the brain electrical signal characteristic index is calculated, and the calculation formula is: Wherein, R represents the brain electrical signal characteristic index, v j represents the voltage of the jth brain electrical signal, d j represents the frequency of the jth brain electrical signal, k j represents the waveform classification coefficient of the jth brain electrical signal, s j represents the spatial energy proportion of the jth brain electrical signal, a j represents the time interval between the jth brain electrical signal and the previous brain electrical signal, and m represents the total number of brain electrical signals collected in the statistical time period;

[0014] According to the brain electrical signal characteristic index, the brain electrical electrode position information is called, the brain electrical signal characteristic index is mapped to the corresponding coordinate position based on the electrode position information, and the real-time brain electrical signal state diagram is drawn.

[0015] Preferably, the acquiring step of the brain electrical signal characteristic analysis result is:

[0016] According to the brain electrical signal real-time state diagram, the brain electrical signal characteristic value is calculated, and the calculation formula is: Wherein, E is the brain electrical signal characteristic value, R is the brain electrical signal characteristic index, u is the real-time average frequency of the brain electrical signal, u0 is the normal average frequency of the brain electrical signal, p is the frequency occupancy rate of the brain electrical signal, and td is the average time delay of the brain electrical signal;

[0017] Based on the brain electrical signal characteristic value, the brain electrical signal characteristic grade is determined, and the brain electrical signal characteristic analysis result is generated.

[0018] Preferably, the acquiring step of the anesthesia state preliminary evaluation result is:

[0019] According to the brain electrical signal characteristic analysis result, the brain electrical signal characteristic mode parameter is called, and the characteristic values between each brain electrical signal characteristic mode are compared one by one, and a brain electrical signal characteristic mode difference combination set is generated;

[0020] Based on the brain electrical signal characteristic mode difference combination set, the brain electrical signal characteristic value is called one by one, the characteristic mode whose characteristic value is lower than the preset characteristic grade standard is determined, the corresponding characteristic mode parameter is marked, and the brain electrical signal characteristic mode limited set is generated by statistics;

[0021] Based on the restricted set of EEG feature patterns, the feature pattern coordinate data of the real-time state diagram of the EEG signal is called, the association relationship between the restricted feature patterns and adjacent feature patterns is determined one by one, an EEG feature pattern association network is established, and a preliminary assessment result of the anesthesia state is generated.

[0022] Preferably, the steps for obtaining the anesthesia state assessment scheme are:

[0023] According to the preliminary assessment results, the anesthesia state assessment index was calculated using the following formula: Among them, Z is the anesthesia state assessment index, G is the number of restricted EEG signal characteristic patterns, T is the duration of characteristic pattern restriction, C is the number of historical abnormalities of the characteristic pattern, and M is the total number of EEG signal characteristic patterns;

[0024] According to the anesthesia state evaluation index, the weight of each characteristic pattern in the preliminary evaluation result is updated one by one, and the calculation is repeated iteratively to generate an anesthesia state evaluation scheme.

[0025] Preferably, the step of obtaining the potential risk status analysis result is:

[0026] Based on the anesthesia state assessment scheme, the EEG signal trajectory data of each characteristic mode in the EEG signal real-time state diagram is called, the degree of deviation of the EEG signal trajectory is calculated and compared with the normal range, the area deviating from the normal trajectory is marked, and a set of abnormal EEG trajectory areas is generated;

[0027] According to the set of abnormal EEG trace areas, the EEG abnormality index is calculated using the following formula: Among them, X is the EEG abnormality index, f r is the average frequency change of EEG signals in the abnormal EEG trace area, e r is the number of energy mutations of EEG signals in the region, t r is the number of abnormal EEG signal time delays in the region, g r is the number of abnormal EEG signal frequencies in the region, h r is the average real-time frequency of EEG signals in the region, and h0 is the normal average frequency of EEG signals in the region;

[0028] Based on the EEG abnormality index, it is determined whether the region is a potential risk state region, and a potential risk state analysis result is generated.

[0029] Preferably, the steps for obtaining the anesthesia risk assessment result are:

[0030] According to the analysis results of the potential risk status, the anesthesia risk occurrence probability index is calculated using the following formula: Wherein, Y is an anesthesia risk occurrence probability index, X is an electroencephalogram abnormality degree index, S is a regional electroencephalogram signal frequency standard deviation, J is a regional electroencephalogram signal abnormal mutation frequency, F is a regional electroencephalogram signal frequency occupancy rate, and U is a regional electroencephalogram signal frequency average value.

[0031] Based on the anesthesia risk occurrence probability index, in combination with the historical anesthesia risk occurrence frequency, the regional anesthesia risk level is divided, and an anesthesia risk assessment result is generated.

[0032] Preferably, the risk warning and control output acquisition step is:

[0033] According to the anesthesia risk assessment result, the relationship between the regional anesthesia risk level and the warning information is matched, the risk warning information content is determined, and a risk warning information content set is generated.

[0034] According to the risk warning information content set, the real-time position information data of the anesthesiologist terminal device is called, the real-time position information data of the terminal device is compared with the position parameters of the risk area in the risk warning information content set, the terminal device that needs to receive the warning information is screened, and a warning information push target set is generated.

[0035] According to the warning information push target set, the risk warning information in the risk warning information content set is pushed to the target terminal device.

[0036] The application also discloses an anesthesia state assessment system based on electroencephalogram data analysis, which comprises the following steps:

[0037] The electroencephalogram data acquisition step: collecting real-time electroencephalogram data, including the voltage, frequency and waveform characteristics of the electroencephalogram signal, calculating the basic characteristic parameters of the electroencephalogram signal according to the real-time electroencephalogram data, and generating an electroencephalogram signal real-time state diagram; based on the electroencephalogram signal real-time state diagram, analyzing the characteristic indexes of the electroencephalogram signal, and generating an electroencephalogram characteristic analysis result;

[0038] The real-time analysis step: based on the electroencephalogram characteristic analysis result, the characteristic mode of the electroencephalogram signal is judged, the anesthesia state preliminary assessment result is acquired, based on the preliminary assessment result, the evaluation model weight is adjusted, and the anesthesia state is accurately evaluated through repeated iteration to generate an anesthesia state evaluation scheme;

[0039] The risk assessment step: from the anesthesia state evaluation scheme, the potential anesthesia risk state is identified, the risk is analyzed, the abnormal electroencephalogram mode is detected, the potential risk state analysis result is acquired, based on the potential risk state analysis result, the anesthesia risk occurrence probability and the estimated influence are evaluated, and an anesthesia risk assessment result is generated.

[0040] Early warning and control step: Based on the anesthesia risk assessment results, risk early warning information is sent to the anesthesiologist to obtain risk early warning and control output.

[0041] Beneficial effects of the present invention: The present invention generates an EEG signal characteristic index and a real-time EEG state diagram by real-time acquisition of the voltage, frequency and waveform characteristic parameters of the EEG signal. On this basis, the anesthetic state is analyzed based on the characteristic index and characteristic level of the EEG signal, the limited anesthetic depth state and the potential risk state are identified, and an anesthetic state assessment scheme based on the EEG characteristic pattern association network is established, so as to accurately assess the anesthetic depth and high-risk state, and improve the anesthetic state assessment capability; in addition, by combining the standard deviation value of the EEG signal frequency in the region, the number of abnormal mutations and the frequency occupancy rate, a more accurate anesthesia risk assessment result is generated, the analysis accuracy of the probability of anesthesia risk occurrence is improved, and the position of the potential risk state is located; and according to the correspondence between the risk level and the warning information, the risk warning information is pushed to the target anesthesiologist, thereby reducing the probability of anesthesia risk occurrence and optimizing anesthesia management, and realizing the active and precise anesthesia state assessment and risk warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 This is a system block diagram of an anesthetic state assessment system based on EEG data analysis according to the present invention;

[0044] Figure 2 Flowchart of the anesthetic state assessment method based on EEG data analysis. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0046] See also Figure 1 As shown, the present invention is an anesthesia state assessment system based on EEG data analysis, which is characterized by comprising:

[0047] The EEG data acquisition module collects real-time EEG data, including the voltage, frequency, and waveform characteristics of the EEG signal, calculates the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generates a real-time state diagram of the EEG signal; based on the real-time EEG state diagram, analyzes the characteristic indicators of the EEG signal and generates EEG characteristic analysis results; the steps for obtaining the real-time state diagram of the EEG signal are as follows:

[0048] Collect the voltage, frequency and waveform characteristics of the patient's real-time EEG signal, determine the waveform classification coefficient based on the waveform characteristics, and form the basic EEG signal data;

[0049] According to the basic EEG signal data, the EEG signal characteristic index is calculated using the following formula: Among them, R represents the EEG signal characteristic index, v j represents the voltage of the jth EEG signal, d j represents the frequency of the jth EEG signal, k j Represents the waveform classification coefficient of the jth EEG signal, s j represents the spatial energy ratio of the jth EEG signal, a j represents the time interval between the jth EEG signal and the previous one, and m represents the total number of EEG signals collected during the statistical time period;

[0050] According to the EEG signal characteristic index, the EEG electrode position information is called, and the EEG signal characteristic index is mapped to the corresponding coordinate position based on the electrode position information to draw a real-time state diagram of the EEG signal.

[0051] The steps for obtaining the EEG feature analysis results are as follows:

[0052] According to the real-time state diagram of the EEG signal, the EEG signal characteristic value is calculated using the following formula: Where E is the EEG signal characteristic value, R is the EEG signal characteristic index, u is the real-time average frequency of the EEG signal, u0 is the normal average frequency of the EEG signal, p is the EEG signal frequency occupancy rate, and td is the average time delay of the EEG signal;

[0053] Based on the EEG signal characteristic value, the EEG signal characteristic level is determined, and an EEG characteristic analysis result is generated.

[0054] The real-time analysis module determines the characteristic pattern of the EEG signal based on the EEG characteristic analysis results and obtains a preliminary evaluation result of the anesthesia state; based on the preliminary evaluation result, the evaluation model weight is adjusted, and the anesthesia state is accurately evaluated by repeated iterations to generate an anesthesia state evaluation plan; the steps for obtaining the preliminary evaluation result of the anesthesia state are as follows:

[0055] According to the electroencephalogram feature analysis result, the electroencephalogram signal feature mode parameter is called, the feature values between each electroencephalogram signal feature mode are compared one by one, and an electroencephalogram signal feature mode difference combination set is generated;

[0056] Based on the electroencephalogram signal feature mode difference combination set, the electroencephalogram signal feature values are called one by one, the feature mode whose electroencephalogram signal feature value is lower than the preset feature level standard is determined, the corresponding feature mode parameter is marked, and an electroencephalogram feature mode restricted set is statistically generated;

[0057] Based on the electroencephalogram feature mode restricted set, the feature mode coordinate data of the electroencephalogram signal real-time state diagram is called, the correlation between the restricted feature mode and the adjacent feature mode is determined one by one, the electroencephalogram feature mode correlation network is established, and a preliminary anesthesia state evaluation result is generated.

[0058] The risk evaluation module identifies a potential anesthesia risk state from the anesthesia state evaluation scheme, analyzes the risk, detects an abnormal electroencephalogram mode, obtains a potential risk state analysis result, evaluates the anesthesia risk occurrence probability and the estimated influence based on the potential risk state analysis result, and generates an anesthesia risk evaluation result; the anesthesia state evaluation scheme is obtained by the following steps:

[0059] According to the preliminary evaluation result, the anesthesia state evaluation index Z is calculated, and the calculation formula is: Wherein, Z is the anesthesia state evaluation index, G is the number of electroencephalogram signal feature mode restrictions, T is the feature mode restriction duration, C is the number of feature mode historical abnormalities, and M is the total number of electroencephalogram signal feature modes;

[0060] The potential risk state analysis result is obtained by the following steps:

[0061] Based on the anesthesia state evaluation scheme, the electroencephalogram signal trajectory data of each feature mode in the electroencephalogram signal real-time state diagram is called, the deviation degree of the electroencephalogram signal trajectory is calculated and compared with the normal range, the area deviating from the normal trajectory is marked, and an abnormal electroencephalogram trajectory area set is generated;

[0062] According to the abnormal electroencephalogram trajectory area set, the electroencephalogram abnormality degree index X is calculated, and the calculation formula is: Wherein, X is the electroencephalogram abnormality degree index, f r is the average frequency change frequency of the electroencephalogram signal in the abnormal electroencephalogram trajectory area, e r is the number of energy mutations of the electroencephalogram signal in the area, t r is the number of time delay abnormalities of the electroencephalogram signal in the area, g r is the number of abnormal frequencies of the electroencephalogram signal in the area, h r is the average real-time frequency of the electroencephalogram signal in the area, and h0 is the normal average frequency of the electroencephalogram signal in the area.

[0063] Based on the EEG abnormality index, it is determined whether the region is a potential risk state region, and a potential risk state analysis result is generated.

[0064] According to the anesthesia state evaluation index, the weight of each characteristic pattern in the preliminary evaluation result is updated one by one, and the calculation is repeated iteratively to generate an anesthesia state evaluation scheme.

[0065] The warning and control module sends risk warning information to the anesthesiologist based on the anesthesia risk assessment results to obtain risk warning and control output. The steps for obtaining the anesthesia risk assessment results are:

[0066] According to the analysis results of the potential risk status, the anesthesia risk occurrence probability index is calculated using the following formula: Among them, Y is the probability index of anesthesia risk, X is the EEG abnormality index, S is the standard deviation of EEG signal frequency in the region, J is the number of abnormal EEG signal mutations in the region, F is the EEG signal frequency occupancy rate in the region, and U is the average EEG signal frequency in the region;

[0067] Based on the anesthesia risk probability index and the number of historical anesthesia risk occurrences, the regional anesthesia risk level is divided and an anesthesia risk assessment result is generated. The steps for obtaining the risk warning and control output are:

[0068] According to the anesthesia risk assessment result, the relationship between the regional anesthesia risk level and the warning information is matched, the risk warning information content is determined, and a risk warning information content set is generated;

[0069] According to the risk warning information content set, the real-time location information data of the anesthesiologist's terminal device is called, and by comparing the real-time location information data of the terminal device with the location parameters of the risk area in the risk warning information content set, the terminal devices that need to receive the warning information are screened, and a warning information push target set is generated;

[0070] According to the warning information push target set, the risk warning information in the risk warning information content set is pushed to the target terminal device.

[0071] See also Figure 2 As shown, the present invention is an evaluation method of an anesthesia state evaluation system based on EEG data analysis, comprising the following steps:

[0072] EEG data acquisition steps: collecting real-time EEG data, including the voltage, frequency, and waveform characteristics of the EEG signal, calculating the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generating a real-time state diagram of the EEG signal; based on the real-time state diagram of the EEG signal, analyzing the characteristic indicators of the EEG signal and generating an EEG characteristic analysis result;

[0073] Real-time analysis step: based on the EEG feature analysis results, determining the characteristic pattern of the EEG signal to obtain a preliminary evaluation result of the anesthesia state; based on the preliminary evaluation result, adjusting the evaluation model weight, repeatedly iterating to accurately evaluate the anesthesia state, and generating an anesthesia state evaluation plan;

[0074] Risk assessment step: identifying potential anesthesia risk states from the anesthesia state assessment scheme, analyzing the risks, detecting abnormal EEG patterns, and obtaining potential risk state analysis results; based on the potential risk state analysis results, assessing the probability of anesthesia risk occurrence and the estimated impact, and generating an anesthesia risk assessment result;

[0075] Early warning and control step: Based on the anesthesia risk assessment results, risk early warning information is sent to the anesthesiologist to obtain risk early warning and control output.

[0076] The present invention discloses the following embodiments:

[0077] The present invention provides a technical solution: an anesthetic state assessment system based on EEG data analysis, comprising:

[0078] EEG data acquisition module

[0079] Collect real-time EEG data, including the voltage, frequency, and waveform characteristics of the EEG signal, calculate the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generate a real-time state diagram of the EEG signal; based on the real-time EEG state diagram, analyze the characteristic indicators of the EEG signal and generate EEG characteristic analysis results.

[0080] Real-time analysis module

[0081] Based on the results of EEG feature analysis, the characteristic pattern of the EEG signal is determined to obtain the preliminary assessment results of the anesthesia state; based on the preliminary assessment results, the evaluation model weights are adjusted, and repeated iterations are performed to accurately assess the anesthesia state and generate an anesthesia state assessment plan.

[0082] Risk Assessment Module

[0083] Identify potential anesthesia risk states from the anesthesia state assessment plan, analyze the risks, detect abnormal EEG patterns, and obtain potential risk state analysis results; based on the potential risk state analysis results, evaluate the probability of anesthesia risk occurrence and the estimated impact, and generate anesthesia risk assessment results.

[0084] Early warning and control module

[0085] Based on the anesthesia risk assessment results, risk warning information is sent to the anesthesiologist to obtain risk warning and control outputs.

[0086] Steps to obtain the real-time status diagram of EEG signals

[0087] The voltage, frequency and waveform characteristics of the patient's real-time EEG signal are collected, and the waveform classification coefficient is determined based on the waveform characteristics to form the basic EEG signal data.

[0088] According to the basic data of EEG signals, the EEG signal characteristic index is calculated using the following formula:

[0089] Among them, R represents the EEG signal characteristic index, v j represents the voltage of the jth EEG signal, d j represents the frequency of the jth EEG signal, k j Represents the waveform classification coefficient of the jth EEG signal, s j represents the spatial energy ratio of the jth EEG signal, a j represents the time interval between the jth EEG signal and the previous EEG signal, and m represents the total number of EEG signals collected during the statistical time period.

[0090] According to the EEG signal characteristic index, the EEG electrode position information is called, and the EEG signal characteristic index is mapped to the corresponding coordinate position based on the electrode position information to draw a real-time status diagram of the EEG signal.

[0091] Specifically, based on the collection of the voltage, frequency and waveform characteristics of each EEG signal, various information of the EEG signal is obtained in real time by installing EEG electrodes and signal amplifiers on the patient's head and setting a fixed sampling frequency.

[0092] The collected real-time frequencies are compared and divided into fixed frequency intervals, for example, 8-13 Hz is used as the alpha wave frequency interval and situations exceeding this interval are recorded.

[0093] The waveform characteristics of the EEG signal, such as α wave, β wave, θ wave, δ wave, etc., are detected according to the signal processing algorithm. When determining the waveform classification coefficient, a previously established waveform classification comparison table is referred to. For example, the α wave classification coefficient is 1.0, the β wave classification coefficient is 2.0, the θ wave classification coefficient is 0.5, the δ wave classification coefficient is 0.2, etc.

[0094] After determining the waveform classification results of each EEG signal, the waveform classification coefficients are uniformly recorded, and then the spatial energy ratio information is summarized. The spatial energy ratio refers to the energy distribution of the EEG signal at different electrode positions, which can be obtained by calculating the ratio of the signal energy at each electrode position to the total energy.

[0095] The above information on voltage, frequency, waveform classification coefficient and spatial energy ratio is then spliced ​​into a unified EEG data record in time sequence to generate complete basic EEG signal data.

[0096] The benefit of the formula is that by introducing multi-dimensional parameters such as the voltage, frequency, waveform classification coefficient, spatial energy ratio and time interval of the EEG signal, it can three-dimensionally depict the dynamic characteristics of the EEG signal, thereby providing more accurate data support for subsequent judgment of the anesthesia status and other derivative analyses.

[0097] v j The steps for obtaining the parameter are as follows: This parameter represents the voltage of the jth EEG signal, which is continuously monitored through EEG electrodes and signal amplifiers. The EEG signal voltage is captured and recorded at intervals of 1ms each time. According to actual measurements, multiple voltage data can be obtained and the average value is used as the final v j .

[0098] d j The steps for obtaining the parameter are as follows: the parameter represents the frequency of the jth EEG signal. First, the EEG signal is subjected to spectrum analysis by the signal processing algorithm to obtain the frequency components of the signal. Then, the frequency is counted at fixed intervals with the statistical time length as the fixed interval, and integrated into the record of each EEG signal to form d j , and its numerical value reflects the frequency characteristics of the EEG signal.

[0099] k j The steps for obtaining the parameter are as follows: This parameter represents the waveform classification coefficient of the jth EEG signal. The value range is set according to the EEG waveform type and the degree of influence of the anesthesia state. In actual monitoring, the waveform recognition algorithm is used to match the pre-determined classification comparison table, and finally k is determined. j is a quantitative value.

[0100] s j The steps for obtaining the parameter are as follows: this parameter represents the spatial energy proportion of the jth EEG signal, which is used to quantify the energy distribution of the EEG signal at different electrode positions. First, the signal energy of each electrode position is calculated, and then compared with the total energy. The energy of each electrode is divided by the total energy to obtain the final ratio, and the value range is between 0.0 and 1.0.

[0101] a j The steps for obtaining the parameter are as follows: this parameter represents the time interval between the jth EEG signal and the previous EEG signal, and is used to characterize the time series characteristics of the EEG signal. The acquisition method is to record the timestamps of adjacent EEG signals and calculate the timestamp difference to obtain the time interval. The smaller the parameter, the shorter the time interval of the EEG signal.

[0102] Throughout the specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0103] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. Anesthesia status assessment system based on EEG data analysis, characterized by: include: The EEG data acquisition module collects real-time EEG data, including the voltage, frequency, and waveform characteristics of the EEG signal, calculates the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generates a real-time state diagram of the EEG signal; Analyzing characteristic indicators of the EEG signal based on the real-time EEG state graph to generate EEG characteristic analysis results; A real-time analysis module, based on the EEG characteristic analysis results, determines the characteristic pattern of the EEG signal and obtains a preliminary evaluation result of the anesthesia state; Based on the preliminary assessment results, adjusting the assessment model weights, iterating repeatedly to accurately assess the anesthesia state, and generating an anesthesia state assessment plan; a risk assessment module, which identifies potential anesthesia risk states from the anesthesia state assessment scheme, analyzes the risks, detects abnormal EEG patterns, and obtains potential risk state analysis results; Based on the potential risk status analysis results, evaluating the probability of occurrence of anesthesia risk and the estimated impact, and generating an anesthesia risk assessment result; The early warning and control module sends risk early warning information to the anesthesiologist based on the anesthesia risk assessment results to obtain risk early warning and control outputs.

2. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the real-time state diagram of the EEG signal are as follows: Collect the voltage, frequency and waveform characteristics of the patient's real-time EEG signal, determine the waveform classification coefficient based on the waveform characteristics, and form the basic EEG signal data; According to the basic EEG signal data, the EEG signal characteristic index is calculated using the following formula: Among them, R represents the EEG signal characteristic index, v j represents the voltage of the jth EEG signal, d j represents the frequency of the jth EEG signal, k j Represents the waveform classification coefficient of the jth EEG signal, s j represents the spatial energy ratio of the jth EEG signal, a j represents the time interval between the jth EEG signal and the previous one, and m represents the total number of EEG signals collected during the statistical time period; According to the EEG signal characteristic index, the EEG electrode position information is called, and the EEG signal characteristic index is mapped to the corresponding coordinate position based on the electrode position information to draw a real-time state diagram of the EEG signal.

3. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the EEG feature analysis results are as follows: According to the real-time state diagram of the EEG signal, the EEG signal characteristic value is calculated using the following formula: Where E is the EEG signal characteristic value, R is the EEG signal characteristic index, u is the real-time average frequency of the EEG signal, u0 is the normal average frequency of the EEG signal, p is the EEG signal frequency occupancy rate, and td is the average time delay of the EEG signal; Based on the EEG signal characteristic value, the EEG signal characteristic level is determined, and an EEG characteristic analysis result is generated.

4. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the preliminary evaluation results of the anesthesia state are: According to the EEG feature analysis result, calling the EEG signal feature pattern parameters, comparing the feature values ​​between each EEG signal feature pattern one by one, and generating an EEG signal feature pattern difference combination set; Based on the EEG signal characteristic pattern difference combination set, calling the EEG signal characteristic values ​​one by one, determining the characteristic patterns whose EEG signal characteristic values ​​are lower than the preset characteristic level standard, marking the corresponding characteristic pattern parameters, and statistically generating the EEG characteristic pattern restricted set; Based on the restricted set of EEG feature patterns, the feature pattern coordinate data of the real-time state diagram of the EEG signal is called, the association relationship between the restricted feature patterns and adjacent feature patterns is determined one by one, an EEG feature pattern association network is established, and a preliminary assessment result of the anesthesia state is generated.

5. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the anesthesia state assessment scheme are: According to the preliminary assessment results, the anesthesia state assessment index was calculated using the following formula: Among them, Z is the anesthesia state assessment index, G is the number of restricted EEG signal characteristic patterns, T is the duration of characteristic pattern restriction, C is the number of historical abnormalities of the characteristic pattern, and M is the total number of EEG signal characteristic patterns; According to the anesthesia state evaluation index, the weight of each characteristic pattern in the preliminary evaluation result is updated one by one, and the calculation is repeated iteratively to generate an anesthesia state evaluation scheme.

6. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the potential risk status analysis result are: Based on the anesthesia state assessment scheme, the EEG signal trajectory data of each characteristic mode in the EEG signal real-time state diagram is called, the degree of deviation of the EEG signal trajectory is calculated and compared with the normal range, the area deviating from the normal trajectory is marked, and a set of abnormal EEG trajectory areas is generated; According to the set of abnormal EEG trace areas, the EEG abnormality index is calculated using the following formula: Among them, X is the EEG abnormality index, f r is the average frequency change of EEG signals in the abnormal EEG trace area, e r is the number of energy mutations of EEG signals in the region, t r is the number of abnormal EEG signal time delays in the region, g r is the number of abnormal EEG signal frequencies in the region, h r is the average real-time frequency of EEG signals in the region, and h0 is the normal average frequency of EEG signals in the region; Based on the EEG abnormality index, it is determined whether the region is a potential risk state region, and a potential risk state analysis result is generated.

7. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the anesthesia risk assessment results are: According to the analysis results of the potential risk status, the anesthesia risk occurrence probability index is calculated using the following formula: Among them, Y is the probability index of anesthesia risk, X is the EEG abnormality index, S is the standard deviation of EEG signal frequency in the region, J is the number of abnormal EEG signal mutations in the region, F is the EEG signal frequency occupancy rate in the region, and U is the average EEG signal frequency in the region; Based on the anesthesia risk occurrence probability index and combined with the historical anesthesia risk occurrence frequency, regional anesthesia risk levels are divided to generate an anesthesia risk assessment result.

8. The anesthesia state assessment system based on EEG data analysis according to claim 1, characterized in that: The steps for obtaining the risk warning and control output are as follows: According to the anesthesia risk assessment result, the relationship between the regional anesthesia risk level and the warning information is matched, the risk warning information content is determined, and a risk warning information content set is generated; According to the risk warning information content set, the real-time location information data of the anesthesiologist's terminal device is called, and by comparing the real-time location information data of the terminal device with the location parameters of the risk area in the risk warning information content set, the terminal devices that need to receive the warning information are screened, and a warning information push target set is generated; According to the warning information push target set, the risk warning information in the risk warning information content set is pushed to the target terminal device.

9. An evaluation method for an anesthetic state evaluation system based on EEG data analysis according to any one of claims 1 to 8, characterized in that: The following steps are involved: EEG data acquisition steps: collect real-time EEG data, including the voltage, frequency and waveform characteristics of the EEG signal, calculate the basic characteristic parameters of the EEG signal based on the real-time EEG data, and generate a real-time state diagram of the EEG signal; Analyzing characteristic indicators of the EEG signal based on the real-time state diagram of the EEG signal to generate an EEG characteristic analysis result; Real-time analysis step: based on the EEG characteristic analysis results, determining the characteristic pattern of the EEG signal and obtaining a preliminary evaluation result of the anesthesia state; Based on the preliminary assessment results, adjusting the assessment model weights, iterating repeatedly to accurately assess the anesthesia state, and generating an anesthesia state assessment plan; Risk assessment step: identifying potential anesthesia risk states from the anesthesia state assessment protocol, analyzing the risks, detecting abnormal EEG patterns, and obtaining potential risk state analysis results; Based on the potential risk status analysis results, evaluating the probability of occurrence of anesthesia risk and the estimated impact, and generating an anesthesia risk assessment result; Early warning and control step: Based on the anesthesia risk assessment results, risk early warning information is sent to the anesthesiologist to obtain risk early warning and control output.

Images