Method and system for evaluating signal quality index SQI in anesthesia equipment

Abstract

The invention provides a method for evaluating a signal quality index SQI in anesthesia equipment, which belongs to the field of anesthesia equipment monitoring and comprises the following steps: S1, performing data preprocessing on electroencephalogram data acquired in a database; s2, performing EEG signal quality judgment on the preprocessed EEG data, including obtaining impedance characteristics, signal spectrum entropy, total power in bandwidth and electrotome interference; s3, iteratively calculating a signal quality index SQI; and S4, when the impedance characteristic weight is less than 90, correcting the signal quality index SQI, reducing the impedance characteristic weight to 0.3, and increasing the electrotome interference weight to 0.6. According to the method, the SQI index oscillation caused by sudden change of a certain type of features can be limited by using two independent features in combination with a real-time correction incremental algorithm, so that the purpose of accurately evaluating the signal quality is achieved, and the influence of disturbance on index data is avoided.

Description

Technical Field

[0001] This invention relates to the field of electroencephalography (EEG) analysis, and in particular to a method and system for evaluating the signal quality index (SQI) in anesthesia equipment. Background Technology

[0002] Currently, analysis based on EEG signal quality assessment has gradually become an important tool in the field of EEG signal quality evaluation in recent years due to its non-invasiveness and real-time nature. However, EEG signal quality assessment still faces many technical challenges. Although the characteristics of the EEG signal itself can directly reflect signal quality, because these characteristics are not independent, in real-time signal quality index algorithms, when the signal encounters accidental short-term perturbations, the index may oscillate. In this case, the evaluation of EEG signal quality based on a single feature will no longer be accurate. Since single-modal acquisition systems are susceptible to noise and interference, traditional EEG signal quality assessment methods are easily affected by perturbations and often fail to accurately evaluate signal quality. Therefore, there is an urgent need for an assessment method that can accurately evaluate EEG signal quality. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention aims to provide a method for evaluating the Signal Quality Index (SQI) in anesthesia equipment. This method assesses the current signal quality by evaluating both the EEG signal itself and its transmission path. Combined with a real-time incremental algorithm, it utilizes four independent features to limit the oscillations in the SQI caused by sudden changes in a certain type of feature, thus achieving an interference-resistant real-time EEG signal quality evaluation algorithm. Based on preprocessing of the EEG signal, this method makes judgments based on four features: impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference. When the signal encounters short-term disturbances, it can avoid oscillations in the SQI, effectively improving the evaluation capability of EEG signal quality.

[0004] Specifically, on the one hand, the present invention provides a method for evaluating the signal quality index (SQI) in anesthesia equipment, which includes the following steps: S1. Perform data preprocessing on the EEG data collected from the database; S2. EEG signal quality assessment of the preprocessed EEG data, including obtaining impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference. Specific steps are: S21. Calculate impedance characteristics, signal spectral entropy, and total power within the bandwidth; S22. Calculate electrosurgical interference using the following formula: ; In the formula, This refers to the energy value of a high-frequency signal, which ranges from 50 to 1000 kHz. This represents the total energy of the signal across the entire frequency band, which ranges from 0.5 to 1000 kHz. This is the current impedance measurement value. This is the impedance baseline value; S3. Iteratively calculate the Signal Quality Index (SQI). The calculation formula is as follows: ; in, Historical weights are used to control the proportion of historical SQI values ​​retained. Incremental weights Used to control the intensity of the increment. The total score is calculated based on four characteristics: impedance, spectral entropy, total power, and electrosurgical interference. S4, when When the value is greater than 90, the signal quality index (SQI) is calculated using the initial weights and saved as the current SQI value; when... When the value is less than 90, the Signal Quality Index (SQI) is corrected by reducing the impedance characteristic weight, increasing the electrosurgical interference weight, and recalculating. and the Signal Quality Index (SQI), if corrected If the value is greater than 90, the corrected signal quality index (SQI) will be used as the current SQI value; if the corrected value is less than 90, the corrected value will be used as the current SQI value. If the value is less than 90, an alarm will be triggered and the anesthesia equipment will be alerted to record that the signal quality index (SQI) at that moment is a non-high-quality signal.

[0005] Preferably, The calculation formula is as follows: ; in, The weights are impedance, spectral entropy, total power, and electrosurgical interference, respectively. The initial weights for the four weights are 0.6, 0.05, 0.05, and 0.3, respectively. When the value is less than 90, the adjusted weights for the four types are 0.3, 0.05, 0.05, and 0.6, respectively. The scores are the satisfaction scores for impedance, spectral entropy, total power, and electrosurgical interference, respectively. is the level adjustment coefficient, and n is the total number of features with a satisfaction score of 1.

[0006] Preferably, when n=4, k n The value is 1.0; when n=3, k n The value is 0.8; when n=2, k n The value is 0.5. When n=2, k n The value is 0.2.

[0007] Preferably, The calculation formulas are as follows: ; ; ; ; in, These are impedance characteristics, signal spectral entropy, and total power, respectively.

[0008] Preferably, the impedance characteristics are calculated in S21 using the following formula: ; ; ; In the formula, the modulation current The impedance information is Im(t), and the modulation voltage signal is V. IM (t), n i and n q These are the co-directional and quadrature components of narrowband noise. This represents the carrier signal obtained during envelope demodulation and secondary modulation. For radians, and These represent different moments before and after.

[0009] Preferably, in S21, the signal spectral entropy is calculated using the following formula: ; ; In the formula, This represents the normalized probability distribution, reflecting the proportion of the nth frequency component in the total energy. It is the power spectral density value of the nth frequency component at time t. The frequency domain value of the nth frequency component, where N / 2 represents the extracted Nyquist frequency domain components. It represents the total energy, specifically the energy level of the nth frequency component.

[0010] Preferably, in step S21, the total power within the bandwidth is calculated using the following formula: ; In the formula, X n (t) represents the value of the nth term after the FFT transformation. Let Y be the complex conjugate form of the signal Xn(t). n (t) represents the power of the corresponding frequency component.

[0011] Preferably, in step S1, a 50Hz notch filter and a 100Hz high-pass filter are cascaded to filter and eliminate the fundamental frequency interference of the electrosurgical unit, retain the main frequency band of the EEG, and extract the energy distribution of the 50-1000kHz frequency band through wavelet transform.

[0012] Preferably, in step S1, the impedance baseline value is updated with a 10-second sliding window.

[0013] On the other hand, the present invention also provides a system for evaluating the signal quality index (SQI) in anesthesia equipment. It includes a data preprocessing unit, a signal quality judgment unit, a signal quality index (SQI) calculation unit, and a signal quality index (SQI) output unit. The data preprocessing unit is used to preprocess the EEG data collected from the database; the signal quality judgment unit is used to judge the EEG signal quality of the preprocessed EEG data, including obtaining impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference; the signal quality index (SQI) calculation unit is used to iteratively calculate the signal quality index (SQI) and correct the signal quality index (SQI) based on the real-time signal quality; the signal quality index (SQI) output unit is used to output a signal quality index (SQI) that meets the requirements to the anesthesia equipment.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The signal quality index (SQI) evaluation method in anesthesia equipment provided by the present invention achieves the purpose of scoring the current signal quality through feature extraction, threshold calculation, and incremental calculation. This method significantly improves the resolution and accuracy of signal quality evaluation, making signal processing more sensitive and avoiding the influence of some interference outside the bandwidth that does not affect signal acquisition on signal quality evaluation, thus preventing distortion of the signal quality index.

[0015] (2) The method for evaluating the Signal Quality Index (SQI) in anesthesia equipment provided by this invention quantifies signal quality from both frequency and impedance perspectives. In terms of frequency, the system condition can be assessed by judging the power frequency ratio (noise), low-frequency drift (artifacts), and spectral entropy (artifacts). The difference between the algorithm and the BIS signal quality index is compared and verified, and the reliability of the signal is determined by reflecting the acquisition link. This algorithm can more accurately evaluate the quality of EEG signals by integrating multiple signal features such as spectrum and time-domain features, avoiding errors that may arise from single-feature evaluation. EEG signals are often interfered with by noise and artifacts such as electromyography (EMG) and eye movements. The method of this invention can improve the accuracy of signal quality assessment through more effective noise suppression techniques.

[0016] (3) The signal quality index (SQI) evaluation system for anesthesia equipment provided by the present invention can be applied to different anesthesia equipment, and is compatible with different EEG signal acquisition equipment. It can also be customized and expanded as needed to meet the needs of different fields such as clinical diagnosis and scientific research. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of the method of the present invention; Figure 2 This is a schematic diagram comparing the evaluation results of the method of this invention and the traditional EEG signal quality assessment algorithm; Figure 3 This is a schematic block diagram of the system of the present invention. Detailed Implementation

[0018] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0019] Firstly, such as Figure 1 As shown, the present invention provides a method for evaluating the signal quality index (SQI) in anesthesia equipment, comprising the following steps: S1. Preprocess the EEG data collected from the database. When anesthesia equipment collects EEG signals, various noises and interferences exist in different frequency bands during the acquisition process. The primary goal of EEG signal evaluation is to avoid the impact of interference outside the bandwidth that does not affect signal acquisition on signal quality assessment. Therefore, real-time low-pass filtering is required, and signal quality is judged based on the signal characteristics within the EEG bandwidth. In a specific embodiment, a cascaded 50Hz notch filter and a 100Hz high-pass filter are used to filter and eliminate electrosurgical fundamental frequency interference, preserving the main EEG frequency band. Wavelet transform is then used to extract the energy distribution in the 50-1000kHz frequency band. In a preferred embodiment, the impedance baseline value is updated with a 10-second sliding window.

[0020] S2. EEG signal quality assessment of the processed EEG data, including impedance characteristics, signal spectral entropy, total power within the bandwidth, and acquisition of electrosurgical interference.

[0021] To separate the impedance channel from the EEG channel and avoid the influence of impedance measurement on the EEG signal, this embodiment uses a high-frequency, weak-current amplitude modulation method. According to Ohm's law, the resulting modulated voltage signal is shown in the following equation: ; In the formula, the modulation current The impedance information is Im(t), and the modulation voltage signal is V. IM (t).

[0022] Based on the premise that signal modulation does not affect EEG acquisition, the algorithm converts the modulated signal into the following equivalent expression: Envelope demodulation yields the following expression, which represents the impedance information, thus enabling the acquisition of the impedance signal, i.e.: ; ; In the formula n i and n q These are the co-directional and quadrature components of narrowband noise. This represents the carrier signal obtained during envelope demodulation and secondary modulation.

[0023] The spectral entropy (SE) of a signal reflects the energy complexity of the signal across various frequency bands in the frequency domain. In algorithms, to extract the power distribution complexity of frequency components within the Nyquist frequency range, the discrete-time spectral entropy (SE) can be described by the following equation: ; ; In the formula, This represents the normalized probability distribution, reflecting the proportion of the nth frequency component in the total energy. It represents the power spectral density value of the nth frequency component at time t, i.e., the energy of that frequency. N / 2 represents the extracted Nyquist frequency domain components. It is the total energy, which is the sum of the power spectral densities of all frequency components.

[0024] The total power within the bandwidth represents the energy characteristics of the signal. In the algorithm, it is obtained by calculating the sum of frequency components within the bandwidth per second of sampling points using a Fast Fourier Transform (FFT), as shown in the following equation: ; In the formula, X n (t) represents the nth value after the FFT transformation, Y n (t) represents the power of the corresponding frequency component. for .

[0025] During calculation, the algorithm performs frequency analysis in n-second windows, with a lower frequency limit of (1 / n) Hz. It's crucial to avoid frequencies below (1 / n) Hz after the Fast Fourier Transform (FFT), as this frequency component may not actually exist in the signal. Secondly, to avoid artifacts caused by discontinuities at the window edges, the window edge data is padded during each calculation.

[0026] For the electrosurgical knife marking, the judgment logic is that when the high-frequency energy ratio is >35% and the impedance change amplitude exceeds the baseline value by 20%, electrosurgical knife interference is triggered, and the electrosurgical knife interference value is recorded as 1.

[0027] ; Where E high E represents the signal energy value in the high-frequency band (50-1000kHz). total Z represents the total energy of the signal across the entire frequency band (0.5-1000kHz). current Z represents the current impedance measurement. baseline This is the impedance baseline value, which in the examples is generally a dynamically updated historical average.

[0028] S3. Iteratively calculate the signal quality index (SQI).

[0029] The relationship between impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference during EEG signal acquisition is analyzed to determine the change in the current signal quality index (SQI) increment. The EEG signal quality is judged by dynamic fusion of the four real-time monitored characteristics.

[0030] Based on the relationships between the above features and signal quality, the core of this method is to determine the change in the current SQI value increment by comparing their threshold relationships. This method collects raw data from a database, performs feature calculations, and then conducts threshold analysis. By iteratively calculating the SQI index, the accuracy of feature extraction is improved, achieving the goal of scoring the current signal quality. Unlike traditional feature extraction algorithms, this algorithm integrates four feature comparison thresholds, calculates the increment, and iterates continuously. The specific process is as follows: Spectral entropy reflects the complexity of the frequency components of the EEG signal during acquisition; that is, the richer the proportion of frequency components in the signal, the higher the spectral entropy. It describes the relative characteristics of signal energy between different frequency bands.

[0031] Impedance values ​​reflect the characteristics of the current acquisition link. Under normal connection conditions, the smaller the impedance value, the better the signal quality. When the contact impedance increases, it will increase Gaussian noise in the signal, low-frequency drift caused by polarization voltage fluctuations, and common-mode noise caused by differential input mismatch, thus degrading the signal quality.

[0032] The total power within the bandwidth reflects the absolute value of the signal energy. During normal acquisition, its magnitude is always within a certain range, which makes up for the limitation of spectral entropy in distinguishing absolute power characteristics.

[0033] Based on the relationship between the above four characteristics and signal quality, the change in the current SQI value increment is determined by comparing their threshold relationships. SQI (Signal Quality Indicator) is a signal quality index that reflects the quality of the signal. In anesthesia depth monitoring equipment, the signal quality index (SQI) is a core technology for ensuring patient safety and precise anesthesia management. Its role far exceeds that of ordinary quality monitoring, directly affecting intraoperative risk control and clinical decision-making. It plays a crucial role in avoiding intraoperative awareness and preventing excessive anesthesia.

[0034] Specifically, the SQI index is calculated as follows: ; Historical weights are used to control the proportion of historical SQI values ​​retained, providing smoothness and resistance to mutations. This controls the strength of the increment. In a specific example, historical weights... The standard value is 0.8, and the incremental weight is 0.2. The threshold comparison results of four independent characteristics—impedance, spectral entropy, total power, and electrosurgical interference—determine the signal quality. Impedance should be below 10kΩ, spectral entropy should be calculated based on the power spectral density (PSD) of the signal and should be within the range [-20, 10], total power should be within the range of 0 to 45Hz, and electrosurgical interference should be below 50kHz. A signal is considered high-quality when all four characteristics are met, acceptable when three characteristics are met, and unusable when fewer than three characteristics are met.

[0035] In a specific embodiment, The calculation formula is as follows: in, The weights are impedance, spectral entropy, total power, and electrosurgical interference, respectively. The initial weights for the four weights are 0.6, 0.05, 0.05, and 0.3, respectively. When the value is less than 90, the four weights after correction are 0.3, 0.05, 0.05, and 0.6, respectively. The scores are the satisfaction scores for impedance, spectral entropy, total power, and electrosurgical interference, respectively. is the level adjustment coefficient, and n is the total number of features with a satisfaction score of 1.

[0036] When n=4, k n The value is 1.0; when n=3, k n The value is 0.8; when n=2, k n The value is 0.5. When n=2, k n The value is 0.2.

[0037] in, The calculation formulas are as follows: in, These are impedance characteristics, signal spectral entropy, and total power, respectively.

[0038] S4, when When the value is greater than 90, the signal quality index (SQI) is calculated using the initial weights (impedance characteristic weight 0.6, electrosurgical interference weight 0.3, signal spectral entropy and total power weight 0.05) and saved as the current SQI value. When the value is less than 90, the Signal Quality Index (SQI) is corrected by reducing the impedance characteristic weight to 0.3 and increasing the electrosurgical interference weight to 0.6, and then recalculating. and the Signal Quality Index (SQI), if corrected If the value is greater than 90, the corrected signal quality index (SQI) will be used as the current SQI value; if the corrected value is less than 90, the corrected value will be used as the current SQI value. If the value is less than 90, an alarm is triggered, and the anesthesia equipment is alerted to record the signal quality index (SQI) at that moment as a non-high-quality signal. Simultaneously, the SQI at that moment is recorded as the current SQI value for subsequent iterative calculations.

[0039] The calculated SQI is a quantitative assessment of the reliability of real-time signals by the system. The reliability of data can be determined by using SQI. The threshold range of the four features is dynamically adjusted according to SQI to form a new threshold, and then SQI is calculated again.

[0040] Traditional methods for calculating the SQI index can cause sudden oscillations when the EEG signal encounters short-term disturbances, rendering the evaluation of EEG signal quality based on a single feature inaccurate. Therefore, the innovation of this method lies in assessing the current signal quality from both the EEG signal itself and the transmission path. Combined with a real-time incremental algorithm, it utilizes four independent features to limit the SQI index oscillations caused by sudden changes in a certain type of feature, thus achieving an interference-resistant real-time EEG signal quality assessment algorithm.

[0041] The calculation process of the SQI index includes feature extraction, threshold calculation, and incremental calculation. By iteratively adjusting the SQI, the purpose of scoring the current signal quality is achieved, avoiding sudden changes in SQI caused by accidental impulse noise.

[0042] During the calculation process, the threshold is adjusted according to the current SQI score to reduce its leniency. As the SQI value increases, the range restricted by the threshold will decrease. Different threshold levels enable the differentiation of signals at different levels.

[0043] In a specific embodiment, experiments were conducted based on an incremental EEG signal quality assessment algorithm. The results showed that in the resting state, the SQI stably recovered to above 90 (out of 100), indicating that the algorithm can effectively identify interference-free states. When blinking or shaking, the SQI dropped to the 0.4-0.6 range, coinciding with manually labeled noisy periods. Traditional methods had a misjudgment rate as high as 42% when electrosurgical intervention was used, while this method reduced the misjudgment rate to 9.7%, and the signal recovery time was shortened from 2.1 seconds in the traditional method to less than 0.8 seconds. Therefore, the goal of accurately assessing EEG signal quality was achieved, avoiding the influence of disturbances on the indicator data.

[0044] The EEG signal quality index test assesses the algorithm's judgment of EEG signal quality by setting different electrode contact methods.

[0045] The experimental equipment in this embodiment was a self-made EEG-fNIRS bimodal brain state monitoring system and a desktop PC. The experimental subjects were healthy adult males. The experiment was conducted indoors. Before the test, sufficient conductive paste was added to the test channel used by the subjects and thoroughly treated with abrasive paste to ensure good electrode contact at the beginning. The signal quality index was tested first with good signal quality (initial condition) for 60s, followed by an increase in artifacts in the electrode signal for 120s, remaining still for 60s, moving the electrode for 60s, and remaining still until the end. The trend of the signal quality index value was observed.

[0046] The algorithm computes data per second, and the experiment lasted 10 minutes, totaling 600 frames of data. Initially, the EEG signal quality was good, and the signal quality index increased. During slight shaking and chewing, electromyographic noise was introduced into the signal, causing the signal quality to deteriorate. The spectral entropy increased significantly, indicating an increase in high-frequency electromyographic components. Abnormal fluctuations in total power within the bandwidth, i.e., sudden changes in the time-domain energy, led to a decrease in the signal quality index. At rest, the signal quality recovered, and the signal quality index increased. Touching the electrodes and significant movements increased low-frequency drift, deteriorating the signal quality. Impedance abrupt changes (>20% of baseline value) were accompanied by low-frequency drift (power increase of 0.5-4Hz), and the signal quality index decreased. At rest, the signal quality index recovered.

[0047] This invention proposes a novel signal quality assessment method. Based on incremental EEG signal quality assessment, this method differs from traditional EEG signal quality assessment methods by introducing an electrosurgical compensation mechanism during the SQI iterative adjustment phase to avoid SQI misjudgments caused by high-frequency electrosurgical pulses. This innovative multi-feature calculation and extraction method, combined with a real-time corrected incremental algorithm, utilizes two independent features to limit SQI index oscillations caused by mutations in a certain type of feature, thus achieving an interference-resistant real-time EEG signal quality assessment algorithm.

[0048] A comparison chart of the results obtained by the method of this invention and conventional calculation is shown below. Figure 2 As shown. Figure 2 Yellow represents the result of the conventional method, and blue represents the result of the method of this application. The results show that, as demonstrated by the above tests, the results calculated using the method of this invention are valid. Figure 2 As shown, EEG data were collected from six participants before surgery. The SQI index was then calculated using the steps described above. The blue curve represents the data obtained from the traditional EEG signal quality assessment algorithm, while the yellow curve represents the data obtained from the incremental EEG signal quality assessment algorithm. Figure 2As can be seen above, when the blue curve exhibits significant oscillations, the yellow curve shows little fluctuation and remains relatively stable. This indicates that traditional EEG signal quality assessment algorithms become inaccurate when encountering disturbances, while the EEG signal quality assessment algorithm of this invention has a significant advantage in anti-interference capabilities.

[0049] Accurate assessment of EEG signal quality: In clinical applications, EEG is commonly used to diagnose neurological diseases such as epilepsy, sleep disorders, brain injury, and cognitive impairment. Low-quality EEG signals may cause artifacts or noise to be misidentified as abnormal brain activity, leading to misdiagnosis or missed diagnosis. Accurate assessment of EEG signal quality allows for the automatic removal of low-quality signals before data analysis, reducing the risk of misdiagnosis and ensuring diagnostic accuracy. EEG signal quality directly affects subsequent signal processing and analysis results. For example, spectral analysis, event-related potential (ERP) analysis, and functional connectivity analysis all rely on high-quality EEG data. EEG signal quality assessment can provide guidance for subsequent signal processing. For instance, when poor signal quality is detected, stronger filtering techniques, artifact removal algorithms, or signal reconstruction methods may need to be applied. Accurate signal quality assessment helps determine suitable signal processing strategies to maximize the recovery of effective brain activity information and avoid interference from invalid signals.

[0050] This method, through precise assessment of EEG signal quality, is crucial for ensuring the reliability and accuracy of EEG data analysis. It not only improves the precision of clinical diagnosis and optimizes data analysis methods, but also reduces the risk of misdiagnosis, increases experimental efficiency, and promotes personalized medicine and neuroscience research. Precise signal quality assessment allows for timely identification and resolution of problems during signal acquisition, ensuring the effectiveness of subsequent analysis and the reliability of the data.

[0051] On the other hand, the present invention also provides a system for evaluating the signal quality index (SQI) in anesthesia equipment, such as... Figure 3 As shown, it includes a data preprocessing unit 1, a signal quality judgment unit 2, a signal quality index (SQI) calculation unit 3, and a signal quality index (SQI) output unit 4. The data preprocessing unit 1 is used to preprocess the EEG data collected from the database; the signal quality judgment unit 3 is used to judge the EEG signal quality of the preprocessed EEG data, including obtaining impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference; the signal quality index (SQI) calculation unit 3 is used to iteratively calculate the signal quality index (SQI) and correct the signal quality index (SQI) according to the real-time signal quality; the signal quality index (SQI) output unit 4 is used to output a signal quality index (SQI) that meets the requirements to the anesthesia equipment.

[0052] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the invention should fall within the protection scope defined by the claims.

Claims

1. A method for evaluating the signal quality index (SQI) in anesthesia equipment, characterized in that: It includes the following steps: S1. Perform data preprocessing on the EEG data collected from the database; S2. EEG signal quality assessment of the preprocessed EEG data, including obtaining impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference. Specific steps are: S21. Calculate impedance characteristics, signal spectral entropy, and total power within the bandwidth; S22. Calculate electrosurgical interference. ； In the formula, This refers to the energy value of a high-frequency signal, which ranges from 50 to 1000 kHz. This represents the total energy of the signal across the entire frequency band, which ranges from 0.5 to 1000 kHz. This is the current impedance measurement value. This is the impedance baseline value; S3. Iteratively calculate the Signal Quality Index (SQI). The calculation formula is as follows: ； in, Historical weights are used to control the proportion of historical SQI values ​​retained. Incremental weights Used to control the intensity of the increment. The total score is calculated based on four characteristics: impedance, spectral entropy, total power, and electrosurgical interference. S4, when When the value is greater than 90, the signal quality index (SQI) is calculated using the initial weights and saved as the current SQI value; when... When the value is less than 90, the Signal Quality Index (SQI) is corrected by reducing the impedance characteristic weight, increasing the electrosurgical interference weight, and recalculating. and the Signal Quality Index (SQI), if corrected If the value is greater than 90, the corrected signal quality index (SQI) will be used as the current SQI value; if the corrected value is less than 90, the corrected value will be used as the current SQI value. If the value is less than 90, an alarm will be triggered and the anesthesia equipment will be alerted to record that the signal quality index (SQI) at that moment is a non-high-quality signal.

2. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: The calculation formula is as follows: ; in, The weights are impedance, spectral entropy, total power, and electrosurgical interference, respectively. The initial weights for the four weights are 0.6, 0.05, 0.05, and 0.3, respectively. When the value is less than 90, the adjusted weights for the four types are 0.3, 0.05, 0.05, and 0.6, respectively. The scores are the satisfaction scores for impedance, spectral entropy, total power, and electrosurgical interference, respectively. is the level adjustment coefficient, and n is the total number of features with a satisfaction score of 1.

3. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: When n=4, k n The value is 1.0; when n=3, k n The value is 0.8; when n=2, k n The value is 0.

5. When n=2, k n The value is 0.

2.

4. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: The calculation formulas are as follows: ； ； ； ； in, These are impedance characteristics, signal spectral entropy, and total power, respectively.

5. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: The impedance characteristic is calculated in S21 using the following formula: ； ； ； In the formula, the modulation current Im(t) represents impedance information, V IM (t) represents the modulation voltage signal, n i and n q These are the in-direction and quadrature components of narrowband noise, respectively. This represents the carrier signal obtained during envelope demodulation and secondary modulation. For radians, and These represent different moments before and after.

6. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: In S21, the signal spectral entropy is calculated using the following formula: ； ； In the formula, This represents the normalized probability distribution, reflecting the proportion of the nth frequency component in the total energy. It is the power spectral density value of the nth frequency component at time t. The frequency domain value of the nth frequency component, where N / 2 represents the extracted Nyquist frequency domain components. It represents the total energy, specifically the energy level of the nth frequency component.

7. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: In step S21, the total power within the bandwidth is calculated using the following formula: ； In the formula, X n (t) represents the value of the nth term after the FFT transformation. Let Y be the complex conjugate form of the signal Xn(t). n (t) represents the power of the corresponding frequency component.

8. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: In step S1, a 50Hz notch filter and a 100Hz high-pass filter are cascaded to filter and eliminate the fundamental frequency interference of the electrosurgical unit, retain the main frequency band of the EEG, and extract the energy distribution of the 50-1000kHz frequency band through wavelet transform.

9. The method for evaluating the signal quality index (SQI) in anesthesia equipment according to claim 1, characterized in that: In step S1, the impedance baseline value is updated with a 10-second sliding window.

10. A system for evaluating the signal quality index (SQI) in an anesthesia device, used in the method for evaluating SQI in any one of claims 1-9, characterized in that: It includes a data preprocessing unit, a signal quality judgment unit, a signal quality index (SQI) calculation unit, and a signal quality index (SQI) output unit. The data preprocessing unit is used to preprocess the EEG data collected from the database; the signal quality judgment unit is used to judge the EEG signal quality of the preprocessed EEG data, including obtaining impedance characteristics, signal spectral entropy, total power within the bandwidth, and electrosurgical interference; the signal quality index (SQI) calculation unit is used to iteratively calculate the signal quality index (SQI) and correct the signal quality index (SQI) based on the real-time signal quality; the signal quality index (SQI) output unit is used to output a signal quality index (SQI) that meets the requirements to the anesthesia equipment.

Images