Impedance respiratory signal filtering method and related components
Patent Information
- Application Number
- CN202610947761.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-29
- Publication Date
- 2026-09-15
Smart Images
Figure CN122744762A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of monitoring equipment technology, and in particular to an impedance respiratory signal filtering method and related components. Background Technology
[0002] Monitoring equipment typically includes impedance respiratory monitoring capabilities to acquire a patient's respiratory status in real time. The standard impedance respiratory signal processing workflow generally involves first performing high-pass and low-pass filtering on the acquired impedance respiratory signal, then applying a respiratory algorithm to process the signal to obtain the respiratory rate. The high-pass filter eliminates the baseline of the impedance respiratory signal, while the low-pass filter eliminates noise and jitter superimposed on it. Since the heart and lungs are interconnected, and respiratory rate is positively correlated with heart rate, current mainstream processing solutions set corresponding low-pass filter cutoff frequencies for different heart rate ranges.
[0003] However, traditional impedance respiratory signal processing schemes have significant limitations in practical applications: when the heart rate is fixed, the cutoff frequency of the low-pass filter is also fixed. For example, if the cutoff frequency of the low-pass filter is low, when the patient's breathing is weak, it may fall below the effective respiratory threshold after low-pass filtering, making it impossible to measure the respiratory rate; conversely, if the cutoff frequency of the low-pass filter is too high, it may introduce more noise interference, which is also detrimental to the calculation of the respiratory rate. Summary of the Invention
[0004] The purpose of this invention is to provide an impedance respiratory signal filtering method and related components to solve the technical problem that the existing technology, due to the binding of the low-pass filter cutoff frequency to the heart rate, cannot adaptively adjust its filtering parameters according to the quality of the impedance respiratory signal, resulting in the possible missed detection of weak breathing or large noise interference.
[0005] To solve the above-mentioned technical problems, the present invention provides an impedance respiratory signal filtering method, comprising: Determine the signal quality of the current impedance respiratory signal, where the signal quality is negatively correlated with the noise in the impedance respiratory signal; Determine the effective respiratory threshold based on the current signal quality; When the respiratory rate cannot be measured based on the effective respiratory threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold.
[0006] Optionally, an effective respiratory threshold may be determined based on the current signal quality, including: Compare the signal quality with the set signal quality threshold; If the signal quality is greater than or equal to the signal quality threshold, the effective breathing threshold is set to the set value. If the signal quality is less than the signal quality threshold, the effective breathing threshold is determined based on the signal quality. The signal quality and the effective breathing threshold are negatively linearly correlated.
[0007] Optionally, before determining the signal quality of the current impedance respiratory signal, the method further includes: When the monitoring function is enabled, obtain the initial heart rate of the monitored object; The cutoff frequency of the low-pass filter is determined based on the initial heart rate.
[0008] Optional, also includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is less than the signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine whether the current signal quality is still below the signal quality threshold; If so, then determine the peak-to-peak value of the current impedance respiratory signal; When the peak-to-peak amplitude exceeds the amplitude threshold, the cutoff frequency of the low-pass filter is reduced by a first preset step size.
[0009] Optional, also includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is not less than the signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine if the current signal quality is less than the signal quality threshold; If not, the cutoff frequency of the low-pass filter is increased by a second preset step size.
[0010] Optionally, determine the signal quality of the current impedance respiratory signal, including: The current impedance respiratory signal is sampled in chronological order to obtain N sampling points, where N is a positive integer not less than 3; Calculate the difference between every two adjacent sampling points to obtain the first-order difference sequence; A polarity sequence is generated based on the positive and negative polarities of the differences in the first-order difference sequence. A difference greater than 0 is denoted as positive polarity, a difference less than 0 is denoted as negative polarity, and a difference equal to 0 retains the polarity of the previous difference. Identify the polarity-switching action in the polarity sequence, and sum the differences of the first-order difference sequences that maintained the same polarity before the current polarity-switching action occurred. If the absolute value of the sum is less than the preset flip threshold, then the polarity flip action is determined to be a valid polarity flip. Determine the effective polarity reversal number Y, where Y is a natural number; The impedance breathing signal noise evaluation coefficient is calculated based on the effective polarity reversal number Y. The impedance breathing signal noise evaluation coefficient is used to characterize the signal quality of the impedance breathing signal, and the magnitude of the impedance breathing signal noise evaluation coefficient is negatively correlated with the signal quality of the impedance breathing signal.
[0011] Optionally, the preset flip threshold is positively correlated with the peak-to-peak amplitude of the current impedance respiratory signal and is not less than the minimum flip threshold.
[0012] Optionally, the impedance respiratory signal noise assessment coefficient is calculated based on the effective polarity reversal number Y, including: The ratio of the effective polarity reversal number Y to the number of upsampling points N is used as the impedance breathing signal noise evaluation coefficient; Alternatively, determine the total number of polarity flips in the polarity sequence; The effective polarity reversal count Y is divided by the total number of polarity reversals as the impedance breathing signal noise evaluation coefficient; Alternatively, the number of noise sampling points can be determined based on the effective polarity reversal number Y; The ratio of the number of noise sampling points to the number of upsampling points N is used as the impedance breathing signal noise evaluation coefficient.
[0013] To address the above problems, the present invention also provides an impedance respiratory signal filtering device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of the impedance breathing signal filtering method described above when executing a computer program.
[0014] To address the aforementioned problems, the present invention also provides a monitoring device, including the impedance respiratory signal filtering device as described above, and further comprising: The signal acquisition component is used to acquire the impedance respiratory signal and heart rate of the monitored subject.
[0015] This invention provides a method for filtering impedance respiratory signals. First, the signal quality of the current impedance respiratory signal is determined. Then, an effective respiratory threshold is determined based on this signal quality. Furthermore, when it is determined that the respiratory rate cannot be measured based on the effective respiratory threshold, the cutoff frequency of the low-pass filter is adaptively adjusted according to the relationship between the current signal quality and the signal quality threshold. Through this linked adjustment mechanism, this invention can adaptively adjust the cutoff frequency of the low-pass filter for different signal qualities, thereby restoring relatively weak impedance respiratory signals, improving detection sensitivity, or filtering out noise to obtain a smoother impedance respiratory signal curve. This invention breaks through the limitation of existing technologies where the low-pass filter cutoff frequency is only tied to heart rate, effectively improving the detection capability of weak respiration and the anti-interference capability in high-noise environments.
[0016] The present invention also provides an impedance respiratory signal filtering device and a monitoring device, both of which have the same beneficial effects as the above-mentioned impedance respiratory signal filtering method. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the prior art and embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A flowchart of an impedance respiratory signal filtering method provided by the present invention; Figure 2 A flowchart of another impedance respiratory signal filtering method provided by the present invention; Figure 3 A flowchart of another impedance respiratory signal filtering method provided by the present invention; Figure 4 A schematic diagram illustrating the mapping relationship between impedance respiratory signal noise evaluation coefficients and effective respiratory thresholds provided by the present invention; Figure 5 This is a schematic diagram illustrating the original polarity reversal of a weak signal. Figure 6 A schematic diagram showing the polarity reversal of a weak signal after fixed filtering. Figure 7 A schematic diagram of polarity reversal after adaptive filtering of a weak signal; Figure 8 This is a schematic diagram illustrating the original polarity reversal of a strong signal. Figure 9 A schematic diagram showing the polarity reversal of a strong signal after fixed filtering. Figure 10 This is a schematic diagram of polarity reversal after adaptive filtering of a strong signal. Figure 11 This invention provides a structural diagram of an impedance respiratory signal filtering device. Figure 12 This is a structural diagram of a monitoring device provided by the present invention. Detailed Implementation
[0019] The core of this invention is to provide an impedance respiratory signal filtering method and related components, which solves the technical problem in the prior art that the low-pass filter cutoff frequency is tied to the heart rate, and the filtering parameters cannot be adaptively adjusted according to the quality of the impedance respiratory signal, resulting in the possible missed detection of weak breathing or large noise interference.
[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] Electromagnetic interference is often present during monitoring, and the impedance respiratory signal acquired by the monitoring equipment may have low quality or contain noise. If traditional impedance respiratory signal filtering methods are used, with a fixed respiratory judgment threshold and a fixed filtering frequency tied to heart rate, noise may be misjudged as breathing when the monitored subject's breathing is weak, or effective breathing may be filtered out, leading to respiratory rate measurement failure or inaccuracy.
[0022] To address the aforementioned technical problems, this invention provides an impedance respiratory signal filtering method. Please refer to [link to relevant documentation]. Figure 1 As shown, Figure 1 The flowchart illustrates an impedance breathing signal filtering method provided by this invention.
[0023] This impedance breathing filter method includes: S11. Determine the signal quality of the current impedance respiratory signal, wherein the signal quality is negatively correlated with the noise in the impedance respiratory signal; First, in this embodiment, after acquiring the current impedance breathing signal, feature extraction and analysis are required to determine the quality of the current impedance breathing signal, including but not limited to the peak-to-peak value and polarity reversal of the signal. It should be noted that signal quality reflects the purity and reliability of the current impedance breathing signal, and it is negatively correlated with the interference in the signal. That is, the more noise such as spikes and baselines contained in the impedance breathing signal, the lower the evaluated signal quality; conversely, the less noise, the higher the signal quality.
[0024] As can be seen, step S11, by determining the signal quality that is negatively correlated with noise, can quantify the signal quality of the current impedance respiratory signal and reflect the severity of the monitoring environment, providing a data basis for further filtering.
[0025] S12. Determine the effective respiratory threshold based on the current signal quality; Secondly, based on the signal quality obtained in step S11, the effective breathing threshold for determining the effective breathing amplitude is determined. Specifically, when the quality of the impedance breathing signal is poor, it indicates that the signal contains a large amount of noise. In order to prevent the interference of high-frequency noise from being misjudged as effective breathing, the effective breathing threshold will be increased accordingly. Conversely, when the signal quality is good, it indicates that the noise interference is small, and a lower set threshold can be maintained or used.
[0026] As can be seen, step S12 uses adaptive determination of the effective breathing threshold based on the current signal quality, which can effectively prevent noise misjudgment caused by the threshold being too low when the signal quality is low, and prevent weak breathing from being missed due to the threshold being too high when the signal quality is high but the breathing is weak, thereby greatly improving the accuracy of human impedance breathing signal monitoring.
[0027] S13. When it is determined that the respiratory rate cannot be measured based on the effective respiratory threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold.
[0028] Finally, based on the effective respiratory threshold determined in step S12, the filtered impedance respiratory signal is adaptively filtered. If the respiratory rate cannot be measured according to the effective respiratory threshold, it indicates that the current filtering parameters do not match the actual impedance respiratory signal. The current signal quality will then be further compared with a preset signal quality threshold, and a targeted adjustment strategy will be implemented. For example, if the current signal quality is good but the respiratory rate cannot be measured, it indicates that the current filtering intensity may be too high, causing weak respiratory signals to be mistakenly filtered out. Therefore, the cutoff frequency of the low-pass filter will be increased, and the filtering intensity will be reduced to increase the probability of detecting weak respiratory signals. Conversely, if the current signal quality is poor and the respiratory rate cannot be measured, it indicates that the effective signal may be overwhelmed by noise. Therefore, the cutoff frequency of the low-pass filter will be reduced to further reduce noise interference and obtain a smoother impedance respiratory signal curve.
[0029] As can be seen, in the absence of a measurable respiratory rate, step S13, by setting an adaptive adjustment mechanism for the low-pass filter cutoff frequency, can raise and lower the low-pass filter cutoff frequency differently for different signal quality. This can not only restore relatively weak impedance respiratory signals and improve detection sensitivity, but also filter out noise to obtain a smoother impedance respiratory signal curve. This completely breaks the limitation of the existing technology where the low-pass filter cutoff frequency is only tied to the heart rate, and effectively improves the detection capability for weak breathing and the anti-interference capability under complex working conditions.
[0030] Furthermore, after completing the adaptive adjustment of the filtering parameters, the monitoring device can use the adjusted effective respiratory threshold and the adjusted low-pass filter cutoff frequency to filter the current impedance respiratory signal. Subsequently, based on the clearer and more matched signal curve obtained after re-filtering, the monitoring device can again perform the determination of effective breathing and the calculation of respiratory rate, and again perform adaptive adjustment and filtering of the filtering parameters until an impedance respiratory signal that meets the calculation requirements is captured.
[0031] Based on the above embodiments, as an optional embodiment, determining the effective respiratory threshold based on the current signal quality includes: Compare the signal quality with the set signal quality threshold; If the signal quality is greater than or equal to the signal quality threshold, the effective breathing threshold is set to the set value. If the signal quality is less than the signal quality threshold, the effective breathing threshold is determined based on the signal quality. The signal quality and the effective breathing threshold are negatively linearly correlated.
[0032] Specifically, this invention pre-sets a signal quality threshold. When the evaluated signal quality is greater than or equal to this threshold, it indicates that the current impedance respiratory signal is relatively pure. In this case, a lower fixed setting value can be directly used as the effective respiratory threshold to avoid excessive computational consumption. Conversely, when the signal quality is less than this threshold, it indicates that the signal has been significantly interfered with. In this case, the effective respiratory threshold increases dynamically in a linear relationship. As the signal quality gradually decreases, the effective respiratory threshold increases linearly, which macroscopically manifests as a negative linear correlation between the effective respiratory threshold and the signal quality.
[0033] As can be seen, this embodiment ensures that the effective breathing threshold is small when the signal quality is good through a linear adjustment mechanism, which significantly improves the detection capability of weak breathing; while when the signal quality deteriorates, the effective breathing threshold increases linearly with the increase of noise, balancing detection sensitivity and anti-interference capability, and effectively improving the accuracy of impedance breathing signal monitoring.
[0034] As an optional embodiment, before determining the signal quality of the current impedance respiratory signal, the method further includes: When the monitoring function is enabled, obtain the initial heart rate of the monitored object; The cutoff frequency of the low-pass filter is determined based on the initial heart rate.
[0035] Specifically, in this embodiment, when the monitoring function is activated, since sufficient continuous respiratory signals have not yet been accumulated to assess signal quality, the initial heart rate of the monitored subject needs to be acquired first. Then, based on the determined initial heart rate, the default initial cutoff frequency of the low-pass filter is determined. Different initial cutoff frequencies are set for different initial heart rate ranges; the higher the heart rate, the higher the cutoff frequency. For example, when the measured initial heart rate is 0-30 beats / minute, the initial cutoff frequency of the low-pass filter can be set to 1Hz; when the initial heart rate is 30-60 beats / minute, the initial cutoff frequency can be set to 1.1Hz, and so on. The specific range division of the initial heart rate and the corresponding default low-pass filter frequency value can be set according to the actual parameters of the device.
[0036] It should be noted that, unlike existing technologies where "once the heart rate is fixed, the low-pass filter frequency is also fixed," in this scheme, the mapping relationship between the initial heart rate and the initial cutoff frequency of the low-pass filter is only used to establish the initial cutoff frequency of the low-pass filter for the first cycle. Once sufficient continuous respiratory signals are obtained to assess signal quality, the dependence on heart rate can be severed, and the process can proceed to the adaptive adjustment of the low-pass filter cutoff frequency based on signal quality and the effective respiratory threshold in the aforementioned S11-S13.
[0037] As can be seen, this embodiment utilizes the mapping relationship between respiratory rate and heart rate to achieve rapid startup of the monitoring equipment and lays a solid foundation for subsequent adaptive dynamic adjustment of the low-pass filter cutoff frequency.
[0038] As an optional embodiment, such as Figure 2 As shown, Figure 2 The flowchart of another impedance respiratory signal filtering method provided by the present invention further includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is less than the signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine whether the current signal quality is still below the signal quality threshold; If so, then determine the peak-to-peak value of the current impedance respiratory signal; When the peak-to-peak amplitude exceeds the amplitude threshold, the cutoff frequency of the low-pass filter is reduced by a first preset step size.
[0039] Specifically, in this embodiment, for impedance respiratory signals with poor signal quality during the initial evaluation, which are below the set signal quality threshold, a frequency reduction and noise reduction cyclic adjustment process is entered when the respiratory rate cannot be measured based on the effective respiratory threshold: First, re-evaluate whether the current signal quality is still below the signal quality threshold. If the current signal quality is no less than the threshold, it means that the signal quality has been restored to a good state after pre-filtering or environmental improvement. Continuing to reduce the frequency may result in the loss of useful signals. Therefore, exit the loop directly and stop adjusting the low-pass filter cutoff frequency. If the current signal quality is still below the signal quality threshold, the cutoff frequency of the low-pass filter needs to be further reduced to enhance its noise reduction capability. However, to prevent excessive frequency reduction from distorting the already weak real breathing signal, a stopping mechanism based on the peak-to-peak amplitude of the impedance breathing signal is introduced. First, the peak-to-peak amplitude of the impedance breathing signal within a fixed time window (e.g., 10 seconds) is determined and compared with a set amplitude threshold (e.g., 0.5mV). If the peak-to-peak amplitude is still greater than the threshold, it indicates that the signal still has sufficient fluctuation energy, allowing the cutoff frequency of the low-pass filter to be reduced by a fixed first preset step size (e.g., 0.05Hz). The first preset step size can be set according to the actual needs of the monitoring equipment. The smaller the value of the first preset step size, the more precise the adjustment of the low-pass filter cutoff frequency, the slower the rate of decrease in the corresponding impedance breathing signal noise evaluation coefficient CV, and the slower the overall adjustment speed. The preferred value range is 0.1Hz to 0.5Hz. Conversely, if the peak-to-peak amplitude has bottomed out to be less than or equal to the amplitude threshold, the frequency reduction adjustment will stop.
[0040] It should be noted that if, during the repeated adjustments of the low-pass filter frequency, the parameters have been adjusted to their limits, but effective respiration still cannot be detected, it indicates that the current signal interference may be too great, or the patient's respiratory signal is too weak to be reconstructed by the algorithm alone. In this case, to ensure the safety of the monitored individual, corresponding abnormal prompts or alarm signals can be output. For example, if the current signal quality is extremely poor and the signal waveform has almost no fluctuations, an alarm can be triggered stating "The monitoring equipment may have detached; please check the monitoring equipment." If the signal quality is relatively good, but the signal waveform has almost no fluctuations, an alarm can be triggered stating "The patient may be experiencing respiratory arrest or suffocation." If interference is detected and impedance respiratory signals are almost undetectable, an alarm can be triggered stating "The current ambient noise is too high." Furthermore, the forms of abnormal prompts or alarm signals include, but are not limited to, text reminders, buzzer warnings, etc. Those skilled in the art can flexibly select appropriate alarm methods according to actual needs, and all should fall within the protection scope of this invention.
[0041] As can be seen, this embodiment addresses situations with significant interference in actual working conditions by setting a dual judgment mechanism that considers both signal quality recovery and peak-to-peak value reaching the minimum. This balances noise reduction and fidelity preservation of impedance breathing signals under complex working conditions, effectively improving the adaptive filtering capability of this invention in complex environments.
[0042] As an optional embodiment, such as Figure 3 As shown, Figure 3 The flowchart of another impedance respiratory signal filtering method provided by the present invention further includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is not less than the signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine if the current signal quality is less than the signal quality threshold; If not, the cutoff frequency of the low-pass filter is increased by a second preset step size.
[0043] Specifically, this embodiment addresses the cyclic adjustment process for impedance respiratory signals with relatively good signal quality during the initial assessment, not less than the signal quality threshold, i.e., when the initial environmental conditions are deemed favorable but the respiratory rate still cannot be measured: First, check again whether the current signal quality is lower than the signal quality threshold. If the current signal quality is already lower than the threshold, it means that increasing the low-pass filter cutoff frequency has introduced significant noise, causing the signal to deteriorate. Therefore, exit the loop directly and stop adjusting the low-pass filter cutoff frequency.
[0044] If the current signal quality is still not less than the signal quality threshold, it indicates that the widening of the frequency band has not introduced noise or the noise is still within an acceptable range. In this case, it is permissible to increase the cutoff frequency of the low-pass filter by a fixed second preset step size (e.g., 0.05Hz) to gradually increase the ability to capture faint breathing. Similarly, the second preset step size can be set according to the actual needs of the monitoring equipment. The smaller the value of the second preset step size, the more precise the adjustment of the cutoff frequency of the low-pass filter, the slower the rate of increase of the corresponding impedance breathing signal noise evaluation coefficient CV, and the slower the increase of the waveform amplitude of the impedance breathing signal. The overall adjustment speed will be slower accordingly. Preferably, the value range of the second preset step size can be 0.1Hz to 0.5Hz.
[0045] As can be seen, this embodiment, targeting the actual working conditions where environmental interference is minimal but breathing is faint, achieves a balance between capturing faint breathing and preventing the introduction of noise by setting a gradually probing and increasing cutoff frequency of the low-pass filter, thus significantly enhancing the sensitivity of the monitoring equipment in capturing faint breathing.
[0046] As an optional embodiment, determining the signal quality of the current impedance respiratory signal includes: The current impedance respiratory signal is sampled in chronological order to obtain N sampling points, where N is a positive integer not less than 3; Calculate the difference between every two adjacent sampling points to obtain the first-order difference sequence; A polarity sequence is generated based on the positive and negative polarities of the differences in the first-order difference sequence. A difference greater than 0 is denoted as positive polarity, a difference less than 0 is denoted as negative polarity, and a difference equal to 0 retains the polarity of the previous difference. Identify the polarity-switching action in the polarity sequence, and sum the differences of the first-order difference sequences that maintained the same polarity before the current polarity-switching action occurred. If the absolute value of the sum is less than the preset flip threshold Sth, then the polarity flip action is determined to be a valid polarity flip. Determine the effective polarity reversal number Y, where Y is a natural number; The impedance breathing signal noise evaluation coefficient CV is calculated based on the effective polarity reversal number Y. The impedance breathing signal noise evaluation coefficient CV is used to characterize the signal quality of the impedance breathing signal, and the magnitude of the impedance breathing signal noise evaluation coefficient CV is negatively correlated with the signal quality of the impedance breathing signal.
[0047] In practical applications, the current impedance respiratory signal can be continuously sampled using a sliding time window. For example, the window length is set to N sampling points, and each time it advances forward with a fixed step value (such as sliding 1 sampling point) (i.e., the first time the sampling points are taken from the 1st to the 10th, the second time the sampling points are taken from the 2nd to the 11th, and so on), thereby realizing real-time dynamic evaluation of the streaming signal.
[0048] Specifically, after acquiring N sampling points within the current window, their first-order difference sequences are calculated and their polarities are determined. For each polarity reversal, the cumulative sum of a continuous first-order difference sequence that maintains the same polarity before the reversal is calculated. This is used to distinguish between normal reversals caused by effective breathing and abnormal reversals caused by noise jitter, where abnormal reversals are recorded as effective reversals. For example, if the extracted first-order difference sequence segment is 12, 2, -2, -1, 6, 18, 7, and two polarity reversals occur, the positive polarity sequence before the first reversal is 12, 2, and its cumulative sum has an absolute value of 14. The negative polarity sequence before the second reversal is "-2, -1", and its cumulative sum has an absolute value of 3. The absolute values mentioned above are compared with a preset flip threshold Sth (e.g., 4): The absolute value of the first flip, 14, is greater than the preset flip threshold Sth, indicating that the polarity flip of the impedance breathing signal has sufficient amplitude change and is a normal fluctuation caused by effective breathing, so it is not included in the count; the absolute value of the second flip, 3, is less than the preset flip threshold Sth, indicating that this may only be a jitter caused by noise, so it is judged as a valid polarity flip representing noise. Based on this, the total number Y of valid polarity flips representing noise jitter in the current N sampling points can be counted.
[0049] Furthermore, the impedance breathing signal noise evaluation coefficient CV can be calculated based on the effective polarity reversal number Y. The impedance breathing signal noise evaluation coefficient CV is obtained by using the formula CV=Y / N. The smaller the value of the impedance breathing signal noise evaluation coefficient CV, the less noise jitter the impedance breathing signal carries, the smoother the signal, and the higher the signal quality. The larger the value of the impedance breathing signal noise evaluation coefficient CV, the more noise jitter the impedance breathing signal carries, the coarser the signal, and the lower the signal quality.
[0050] Furthermore, this embodiment provides a method for real-time adjustment of the effective respiratory threshold Rth based on the impedance respiratory signal noise evaluation coefficient CV, such as... Figure 4 As shown, Figure 4 This invention provides a schematic diagram illustrating the mapping relationship between impedance respiratory signal noise evaluation coefficients and effective respiratory thresholds. As mentioned above, a minimum value R of the effective respiratory threshold can be preset. min Maximum value R max The impedance respiratory signal noise evaluation coefficient threshold Cth corresponds to the signal quality threshold. When the calculated impedance respiratory signal noise evaluation coefficient CV is less than or equal to the impedance respiratory signal noise evaluation coefficient threshold Cth, it indicates that the signal quality is high, and the effective respiratory threshold Rth takes the minimum value R. min When the impedance respiratory signal noise evaluation coefficient CV is greater than the impedance respiratory signal noise evaluation coefficient threshold Cth, it indicates low signal quality. Rth increases linearly with the increase of the impedance respiratory signal noise evaluation coefficient CV. For example, as the impedance respiratory signal noise evaluation coefficient CV gradually increases from the impedance respiratory signal noise evaluation coefficient threshold Cth to 1, the effective respiratory threshold Rth decreases from its minimum value R. min linearly increases to the maximum value R max Through this design, the present invention can achieve the aforementioned goal of having a smaller effective breathing threshold when the signal quality is good, thereby improving the detection capability of weak breathing, and having a larger effective breathing threshold when the signal quality deteriorates, thus preventing noise from being misidentified as breathing.
[0051] The threshold value Cth of the impedance respiratory signal noise evaluation coefficient is related to the noise tolerance of the respiratory detection algorithm. Each monitoring device uses a different respiratory detection algorithm, resulting in varying noise tolerances for the input respiratory waveform. As a preferred calibration method, since weak signals are more easily affected by noise and are more sensitive to the CV value, the effective respiratory threshold can be set to the minimum value R during calibration. min The amplitude is slightly greater than R. minFurthermore, when respiratory waveforms with different signal quality are input into the respiratory detection algorithm, the expected respiratory rate will theoretically become worse as the waveform with high signal quality is input to the waveform with low signal quality. The critical waveform at which the respiratory detection algorithm can output an accurate respiratory rate is recorded, and the impedance respiratory signal noise evaluation coefficient corresponding to the critical waveform is used as Cth.
[0052] As can be seen, this embodiment separates the effective respiratory waveform from high-frequency noise by using a first-order differential sequence and amplitude determination before polarity reversal, thereby realizing the digital representation of complex impedance respiratory signals, saving the computing resources of the monitoring equipment, and providing a basis for the adaptive adjustment of the subsequent low-pass filter.
[0053] As an optional embodiment, the preset flip threshold Sth is positively correlated with the peak-to-peak amplitude of the current impedance respiratory signal and is not less than the minimum limit value S. min .
[0054] Specifically, when determining the preset flip threshold Sth, the peak-to-peak value Pth of N sampling points within the current sliding window is calculated to characterize the current overall respiratory amplitude. Subsequently, the preset flip threshold Sth can be adaptively set with reference to this peak-to-peak value Pth; for example, based on experience, the ratio formula can preferably be set as Sth = Pth / 5. This positive correlation mechanism ensures that regardless of the depth of the patient's respiratory amplitude, the noise assessment scale can be adaptively adjusted according to the preset ratio. Simultaneously, when the sliding window contains only noise, Pth itself is small, and calculating Sth from Pth is meaningless; therefore, a minimum limit value Sth needs to be set. min When Pth / 5 is calculated min When, then take Sth = S min .
[0055] As can be seen, this embodiment further improves the anti-interference capability of the impedance breathing signal noise evaluation coefficient CV calculation by limiting the preset flip threshold Sth, ensuring that objective and accurate signal quality evaluation results can be output under various real-world working conditions.
[0056] In addition, the minimum value R of the above-mentioned effective respiratory threshold min Maximum value R max And the minimum limit value S of the preset flip threshold Sth min It can be determined based on the inherent physical characteristics of the monitoring equipment and experimental calibration. Specifically, S min Related to the inherent noise of each monitoring device, existing impedance respiration measurement principles all involve applying a weak current between measuring electrodes. When this current passes through the human chest cavity, it generates a voltage change at the same frequency as the human respiratory rate. However, the noise current of the monitoring device is also conducted to the electrodes, thus generating a noise voltage. With the inherent noise of the monitoring device remaining constant, the greater the impedance of the human chest, the greater the generated noise voltage. Based on the physiological characteristic that the impedance of the human chest is generally below 2000Ω, a 2000Ω resistor is shorted between the measuring electrodes. The peak-to-peak noise amplitude Pth of the monitoring device under the limiting physiological impedance condition is measured, and a minimum limiting value S is set. min Set to a value no less than the peak-to-peak noise amplitude under this condition.
[0057] The minimum effective respiratory threshold R can be set according to actual needs. min Take S min Two to three times the normal volume of breathing, to avoid noise being identified as effective breathing.
[0058] In addition, the maximum value R of the effective respiratory threshold max Set to R min 1.5 to 2 times; R max When used in conjunction with subsequent respiration detection algorithms, it can be used to simulate experiments to evaluate R... max Perform parameter verification, such as setting R according to actual needs. max For R min The value is 1.5 times that of the impedance respiratory signal noise evaluation coefficient. This verifies whether the subsequent respiratory detection algorithm can accurately measure the respiratory rate under respiratory waves of different amplitudes when the noise evaluation coefficient is close to 1, and then increases R sequentially. max (e.g. R) max =1.6 R min R max = 1.7 R min ...) ratios were verified and compared based on actual needs and by comparing different R... max To obtain a suitable value for R. max The larger the value, the lower the probability of misjudging noise as valid breathing, but the weaker the ability to detect faint breathing.
[0059] As an optional embodiment, the impedance respiratory signal noise evaluation coefficient CV is calculated based on the effective polarity reversal number Y, including: The ratio of the effective polarity reversal number Y to the number of upsampling points N is used as the impedance breathing signal noise evaluation coefficient CV; Alternatively, determine the total number of polarity flips in the polarity sequence; The effective polarity reversal count Y is divided by the total number of polarity reversals as the impedance breathing signal noise evaluation coefficient CV; Alternatively, the number of noise sampling points can be determined based on the effective polarity reversal number Y; The ratio of the number of noise sampling points to the number of upsampling points N is used as the impedance breathing signal noise evaluation coefficient CV.
[0060] Specifically, this embodiment provides three parallel methods for calculating the impedance respiratory signal noise evaluation coefficient (CV), quantifying the noise pollution level of the impedance respiratory signal from different dimensions. In practical applications, those skilled in the art can choose one or more of these methods based on the performance differences of processors in different models of monitoring devices and different clinical monitoring accuracy requirements. It should be noted that these three calculation methods are merely specific implementation forms. Any other logical calculation method that utilizes similar ideas, i.e., calculating the impedance respiratory signal noise evaluation coefficient (CV) through statistical characteristics to characterize signal quality, should fall within the protection scope of this invention, and will not be elaborated further here.
[0061] The first method involves dividing the effective polarity reversal number Y, representing noise jitter, by the total number of sampling points N within the sliding window, and using the ratio as the impedance breathing signal noise evaluation coefficient CV. Compared to traditional spectral analysis methods such as Fourier transform, this method has the advantages of simple computational logic, low resource consumption, and no need for redundant feature statistics, making it suitable for scenarios where the monitoring equipment has limited computing power and low requirements for monitoring accuracy.
[0062] The second approach involves simultaneously calculating the total number of polarity switches occurring within the time window during first-order differential polarity analysis. Finally, the proportion of effective polarity reversals to the total number of polarity reversals is calculated as the impedance breathing signal noise evaluation coefficient (CV). This method can eliminate interference to a certain extent and has strong anti-interference capabilities, making it suitable for scenarios with significant interference in actual operating conditions.
[0063] The third method marks the sampling points covered during effective polarity reversal as noise sampling points, then counts the total number of noise sampling points within the entire window, and finally calculates its proportion in the total sampling points as the impedance breathing signal noise evaluation coefficient (CV). This method calculates the impedance breathing signal noise evaluation coefficient (CV), which characterizes the signal quality, based on the duration of each noise event. The time-based calculation makes the results more accurate, but the corresponding computational requirements are also relatively high. It is suitable for monitoring scenarios with high requirements for anti-interference capabilities and data accuracy.
[0064] As can be seen, by providing three different impedance respiratory signal noise evaluation coefficient CV calculation formulas, this embodiment enables the invention to be flexibly adapted to various high-end and low-end hardware platforms and various monitoring scenarios.
[0065] As a more specific embodiment, please refer to Figure 5 , Figure 6and Figure 7 As shown, Figure 5 This is a schematic diagram illustrating the original polarity reversal of a weak signal. Figure 6 This is a schematic diagram illustrating the polarity reversal of a weak signal after fixed filtering. Figure 7 This is a schematic diagram of the polarity reversal after adaptive filtering of a weak signal.
[0066] This embodiment uses a set of impedance respiratory signal data acquired at a sampling rate of 100Hz for 10 seconds. The preset threshold for the impedance respiratory signal noise evaluation coefficient (i.e., the corresponding signal quality threshold) is 0.02. As can be seen from the original signal, this impedance respiratory signal is extremely weak and accompanied by significant noise. Taking the first calculation method as an example, directly calculating its first-order differential polarity yields an impedance respiratory signal noise evaluation coefficient CV of 0.131 (i.e., 131 effective polarity flips occurred in 1000 sampled data points), far exceeding the threshold of 0.02. If a fixed-frequency filter is used, such as setting a window moving average filter time of 0.8 seconds, corresponding to a lower cutoff frequency, although the filtered CV value drops to 0.003, the amplitude of the originally weak true respiratory signal is significantly reduced, which may lead to missed detection in subsequent judgments due to the signal falling below the effective respiratory threshold.
[0067] Using the adaptive adjustment mechanism provided by this invention, taking the first calculation method as an example, since the detected value of the impedance respiratory signal noise evaluation coefficient CV is 0.003, which is much smaller than the threshold of 0.02, the cutoff frequency of the low-pass filter is gradually increased with a fixed second preset step size. For example, the window moving average filtering time is reduced from 0.8 seconds to 0.3 seconds. After adjustment, the CV value of the signal rises to 0.01, still less than the threshold of 0.02, but the amplitude of the impedance respiratory signal is significantly improved. If necessary, the cutoff frequency can be further increased to continue to enhance the signal amplitude, thereby achieving maximum restoration of weak respiratory signals while ensuring low noise.
[0068] As another specific embodiment, please refer to Figure 8 , Figure 9 and Figure 10 As shown, Figure 8 This is a schematic diagram illustrating the original polarity reversal of a strong signal. Figure 9 This is a schematic diagram of the polarity reversal after a strong signal has been fixedly filtered. Figure 10 This is a schematic diagram of the polarity reversal after adaptive filtering of a strong signal.
[0069] This embodiment uses another set of impedance respiratory signal data acquired at a sampling rate of 100Hz for 20 seconds. The preset threshold for the impedance respiratory signal noise evaluation coefficient (i.e., the corresponding signal quality threshold) is 0.02. Taking the first calculation method as an example, the impedance respiratory signal itself has a strong amplitude, but it is mixed with dense strong noise. The directly calculated impedance respiratory signal noise evaluation coefficient CV is 0.150 (i.e., there are 300 valid polarity flips in 2000 sampled data). The first filtering is performed with a window moving average filtering time of 0.3 seconds, corresponding to a high cutoff frequency. The signal amplitude is still strong enough after filtering, but the impedance respiratory signal noise evaluation coefficient CV is 0.037, which is still greater than the set threshold of 0.02. This indicates that the high-frequency noise has not been filtered out completely, which can easily lead to misjudgment.
[0070] In this case, using the adaptive adjustment method of the present invention, if the respiratory rate cannot be accurately measured and the current impedance respiratory signal noise evaluation coefficient CV is greater than the threshold, and the signal amplitude is still large, the cutoff frequency of the low-pass filter is reduced to enhance noise reduction (i.e., the window moving average filtering time is increased from 0.3 seconds to 1.0 seconds). After adjustment, the impedance respiratory signal noise evaluation coefficient CV is reduced to 0.008, and the signal curve becomes smoother. It is evident that the adaptive filtering method provided by the present invention can successfully isolate strong noise interference, greatly improving the accuracy of respiratory rate calculation under complex operating conditions.
[0071] To address the above problems, the present invention also provides an impedance respiratory signal filtering device, such as... Figure 11 As shown, Figure 11 A structural diagram of an impedance respiratory signal filtering device provided by the present invention. The impedance respiratory signal filtering device includes: Memory 41 is used to store computer programs; Processor 42 is used to implement the steps of any of the impedance breathing signal filtering methods described above when executing a computer program.
[0072] For an introduction to the impedance respiratory signal filtering device provided by the present invention, please refer to the embodiments of the impedance respiratory signal filtering method described above, which will not be repeated here.
[0073] To address the above problems, the present invention also provides a monitoring device, such as... Figure 12 As shown, Figure 12 A structural diagram of a monitoring device provided by the present invention. The monitoring device includes the impedance respiratory signal filtering device as described above, and further includes: The signal acquisition component 43 is used to acquire the impedance respiratory signal and heart rate of the monitored object.
[0074] For a description of the monitoring device provided by this invention, please refer to the above-described embodiment of the impedance respiratory signal filtering method, which will not be repeated here.
[0075] It should be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0076] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. An impedance respiratory signal filtering method, characterized in that, include: Determine the signal quality of the current impedance respiratory signal, wherein the signal quality is negatively correlated with the noise in the impedance respiratory signal; Determine the effective respiratory threshold based on the current signal quality; When it is determined that the respiratory rate cannot be measured based on the effective respiratory threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold.
2. The impedance respiratory signal filtering method as described in claim 1, characterized in that, Determining an effective respiratory threshold based on the current signal quality includes: The signal quality is compared with a set signal quality threshold. If the signal quality is greater than or equal to the signal quality threshold, then the effective breathing threshold is set to a set value; If the signal quality is less than the signal quality threshold, then the effective breathing threshold is determined based on the signal quality, and the signal quality is negatively linearly correlated with the effective breathing threshold.
3. The impedance respiratory signal filtering method as described in claim 1, characterized in that, Before determining the signal quality of the current impedance respiratory signal, the following steps are also included: When the monitoring function is enabled, obtain the initial heart rate of the monitored object; The cutoff frequency of the low-pass filter is determined based on the initial heart rate.
4. The impedance respiratory signal filtering method as described in claim 1, characterized in that, Also includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is less than a signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine whether the current signal quality is still less than the signal quality threshold; If so, then determine the peak-to-peak value of the current impedance respiratory signal; When the peak-to-peak value of the amplitude is greater than the amplitude threshold, the cutoff frequency of the low-pass filter is reduced by a first preset step size.
5. The impedance respiratory signal filtering method as described in claim 1, characterized in that, Also includes: Determine whether the signal quality of the first determined impedance respiratory signal is less than the signal quality threshold; When the signal quality of the initially determined impedance respiratory signal is not less than a signal quality threshold, the cutoff frequency of the low-pass filter is adjusted according to the relationship between the current signal quality and the signal quality threshold, including: Determine whether the current signal quality is less than the signal quality threshold; If not, the cutoff frequency of the low-pass filter is increased by a second preset step size.
6. The impedance respiratory signal filtering method according to any one of claims 1 to 5, characterized in that, Determine the signal quality of the current impedance respiratory signal, including: The current impedance respiratory signal is sampled in chronological order to obtain N sampling points, where N is a positive integer not less than 3; Calculate the difference between every two adjacent sampling points to obtain a first-order difference sequence; A polarity sequence is generated based on the positive and negative polarities of the differences in the first-order difference sequence, wherein a difference greater than 0 is recorded as positive polarity, a difference less than 0 is recorded as negative polarity, and a difference equal to 0 retains the polarity of the previous difference. Determine the polarity reversal action in the polarity sequence where the polarity changes, and sum the differences of the first-order difference sequences that maintain the same polarity before the current polarity reversal action occurs; If the absolute value of the sum is less than the preset flip threshold, then the polarity flip action is determined to be a valid polarity flip. Determine the effective polarity reversal number Y, where Y is a natural number; The impedance breathing signal noise evaluation coefficient is calculated based on the effective polarity reversal number Y. The impedance breathing signal noise evaluation coefficient is used to characterize the signal quality of the impedance breathing signal, and the magnitude of the impedance breathing signal noise evaluation coefficient is negatively correlated with the signal quality of the impedance breathing signal.
7. The impedance respiratory signal filtering method as described in claim 6, characterized in that, The preset flip threshold is positively correlated with the peak-to-peak amplitude of the current impedance respiratory signal and is not less than the minimum flip threshold.
8. The impedance respiratory signal filtering method as described in claim 6, characterized in that, The impedance respiratory signal noise evaluation coefficient is calculated based on the effective polarity reversal number Y, including: The ratio of the effective polarity reversal number Y to the number of sampling points N is used as the impedance breathing signal noise evaluation coefficient; Alternatively, determine the total number of polarity flips in the polarity sequence; The effective polarity reversal number Y divided by the total number of polarity reversals is used as the impedance breathing signal noise evaluation coefficient; Alternatively, the number of noise sampling points can be determined based on the effective polarity reversal number Y; The ratio of the number of noise sampling points to the number of sampling points N is used as the impedance breathing signal noise evaluation coefficient.
9. An impedance respiratory signal filtering device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the impedance respiratory signal filtering method as described in any one of claims 1 to 8 when executing the computer program.
10. A monitoring device, characterized in that, Including the impedance respiratory signal filtering device as described in claim 9, further comprising: The signal acquisition component is used to acquire the impedance respiratory signal and heart rate of the monitored subject.