WO2018201683A1 - 生理信号的同源性识别方法及装置 - Google Patents

生理信号的同源性识别方法及装置 Download PDF

Info

Publication number
WO2018201683A1
WO2018201683A1 PCT/CN2017/110350 CN2017110350W WO2018201683A1 WO 2018201683 A1 WO2018201683 A1 WO 2018201683A1 CN 2017110350 W CN2017110350 W CN 2017110350W WO 2018201683 A1 WO2018201683 A1 WO 2018201683A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
physiological signals
data
physiological
different kinds
Prior art date
Application number
PCT/CN2017/110350
Other languages
English (en)
French (fr)
Inventor
刘三超
苏健伟
孙泽辉
孙白雷
叶文宇
刘立汉
李明
Original Assignee
深圳迈瑞生物医疗电子股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳迈瑞生物医疗电子股份有限公司 filed Critical 深圳迈瑞生物医疗电子股份有限公司
Priority to CN201780086365.2A priority Critical patent/CN110312466A/zh
Publication of WO2018201683A1 publication Critical patent/WO2018201683A1/zh
Priority to US16/672,640 priority patent/US11457873B2/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7246—Details of waveform analysis using correlation, e.g. template matching or determination of similarity
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7253—Details of waveform analysis characterised by using transforms
    • A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/021—Measuring pressure in heart or blood vessels
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
    • A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
    • A61B5/14551—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters for measuring blood gases
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]

Definitions

  • the invention relates to the technical field of medical instrument control, in particular to a method and device for identifying homology of physiological signals.
  • the function of the monitor for detecting various physiological conditions of patients has also evolved from the monitoring of single physiological signals to the monitoring of various physiological signals, and the analysis of physiological signals of patients.
  • the method has also evolved from single signal analysis to multi-parameter fusion analysis to improve the reliability and accuracy of the analysis results. That is, the same monitor can monitor a variety of physiological signals of the patient, for example, multiple parameters such as an electrocardiogram signal, a blood pressure, a blood oxygen signal, and the like can be simultaneously detected, and the observed data of the plurality of parameters obtained can be obtained.
  • the analysis and judgment of the user's current physical state improves the reliability and accuracy of the analysis result compared to the analysis of the current single physiological signal observation result of the user.
  • monitor can simultaneously monitor multiple physiological parameters of the patient and perform fusion analysis, it reduces the false alarm or false negative of the monitor.
  • monitor A is used to monitor patient B's ECG signal
  • patient C's blood pressure and blood oxygen signal are simultaneously monitored.
  • monitor A is used to monitor patient B's ECG signal
  • patient C's blood pressure and blood oxygen signal are simultaneously monitored.
  • the existing monitor it is not possible to distinguish whether the multiple signals currently detected are from the same patient.
  • multi-parameter fusion analysis only all the signals detected by the current instrument are analyzed. It does not distinguish between patients from different signals.
  • the monitor will analyze the different types of physiological signals of multiple patients during the fusion analysis because of the various psychological signals for fusion analysis. From different patients, and the specific conditions of each patient are different, which will lead to serious false positives or false negatives in the results of the analysis, resulting in a significant decline in the credibility of multi-parameter fusion analysis.
  • a method for identifying homology of physiological signals comprising:
  • the two different kinds of physiological signals are determined to be homologous.
  • the calculating the joint feature between the signal data of the two different kinds of physiological signals according to the waveform matching information comprises:
  • the method further includes:
  • the signal data of the two different kinds of physiological signals are subjected to high-pass filtering processing and low-pass filtering processing.
  • the method after receiving the signal data of the two different kinds of physiological signals, the method further includes:
  • the signal data of the two different kinds of physiological signals are waveform matched.
  • the extracting the feature data of the signal data of the two different kinds of physiological signals separately includes:
  • the prompt information of the signal abnormality is generated and the user is prompted.
  • the method before performing waveform matching on the signal data of the two different kinds of physiological signals, the method further includes:
  • the signal data is waveform matched; if not, the signal data for receiving the two different kinds of physiological signals is continued.
  • the calculating the homologous reference coefficients corresponding to the two different kinds of physiological signals according to the joint feature further includes:
  • the calculating the homologous reference coefficients corresponding to the two different kinds of physiological signals according to the joint feature further includes:
  • the mean or mean square error of the difference sequence is calculated as a homologous reference coefficient corresponding to the two different kinds of physiological signals.
  • the method before performing waveform matching on the signal data of the two different kinds of physiological signals, the method further includes:
  • the waveform matching of the signal data of the two different kinds of physiological signals is specifically:
  • Waveform matching is performed on the converted signal data corresponding to the signal data of the two different kinds of physiological signals.
  • a homology identification device for physiological signals comprising:
  • a signal data receiving module configured to receive signal data of two different kinds of physiological signals
  • a waveform matching module configured to perform waveform matching on signal data of the two different kinds of physiological signals, and determine waveform matching information in signal data of the two physiological signals;
  • a joint feature acquiring module configured to calculate a joint feature between the signal data of the two different kinds of physiological signals according to the waveform matching information
  • a homology reference coefficient calculation module configured to calculate a homologous reference coefficient corresponding to the two different kinds of physiological signals according to the joint feature
  • the homology identification module is configured to determine that the two different kinds of physiological signals are homologous if the homologous reference coefficient is greater than a preset value.
  • the joint feature acquiring module is further configured to determine a matching peak in the two different kinds of physiological signals according to the waveform matching information, and calculate the two different kinds of physiological A sequence of differences of time points corresponding to the matched peaks in the signal, the difference sequence being used as a joint feature between the signal data of the two different kinds of physiological signals.
  • the apparatus further includes a data pre-processing module, configured to perform high-pass filtering processing and low-pass filtering processing on the signal data of the two different kinds of physiological signals.
  • a data pre-processing module configured to perform high-pass filtering processing and low-pass filtering processing on the signal data of the two different kinds of physiological signals.
  • the apparatus further includes a signal quality parameter calculation module, configured to separately extract feature data of signal data of the two different kinds of physiological signals; and calculate according to preset signal quality parameters.
  • the formula calculates a signal quality parameter corresponding to the feature data; if the signal quality parameter does not satisfy the preset signal quality parameter threshold, determining that the signal data of the physiological signal corresponding to the signal quality parameter is invalid And invoking the waveform matching module if the signal quality parameter satisfies a preset signal quality parameter threshold.
  • the apparatus further includes a single signal parameter calculation module, configured to determine a signal type of the physiological signal for any kind of physiological signal, and determine a single signal parameter corresponding to the signal type.
  • a single signal parameter calculation module configured to determine a signal type of the physiological signal for any kind of physiological signal, and determine a single signal parameter corresponding to the signal type.
  • Type calculating a single signal parameter corresponding to the physiological signal according to a preset single signal parameter calculation formula and the characteristic data; determining whether the single signal parameter satisfies a preset single signal parameter threshold; and the single signal parameter Performing waveform matching on the signal data of the two different kinds of physiological signals when the preset single signal parameter threshold is met; generating a signal abnormality when the single signal parameter does not satisfy the preset single signal parameter threshold
  • the device further includes a buffer filling module, configured to determine, according to signal data of the two different kinds of physiological signals, whether the data size of the signal data is greater than or equal to a preset data amount threshold, where the waveform matching module is invoked if the data amount of the signal data is greater than or equal to a preset data amount threshold; the data amount of the signal data is smaller than the preset data.
  • a buffer filling module configured to determine, according to signal data of the two different kinds of physiological signals, whether the data size of the signal data is greater than or equal to a preset data amount threshold, where the waveform matching module is invoked if the data amount of the signal data is greater than or equal to a preset data amount threshold; the data amount of the signal data is smaller than the preset data.
  • the signal data receiving module is called.
  • the homology reference coefficient calculation module is further configured to calculate a physiological signal with the two different kinds according to a preset homology reference coefficient calculation formula and the difference sequence. Corresponding homologous reference coefficients.
  • the homology reference coefficient calculation module is further configured to calculate an average value or a mean square error of the difference sequence as a homologous reference corresponding to the two different kinds of physiological signals. coefficient.
  • the device further includes a signal data transformation module, configured to separately perform Fourier transform on the signal data of the two different kinds of physiological signals to obtain transformed signal data;
  • the waveform matching module is further configured to perform waveform matching on the transformed signal data corresponding to the signal data of the two different kinds of physiological signals.
  • the homology identification method and apparatus of the above physiological signals After the homology identification method and apparatus of the above physiological signals are used, after the monitor monitors signal data of a plurality of different kinds of physiological signals, the joint characteristics of the signal data of the monitored different kinds of physiological signals are adopted.
  • the extraction and calculations determine the size of the homologous reference parameter that can identify the likelihood that two different kinds of physiological signals are homologous, thereby determining whether the two physiological signals are homologous.
  • the homology analysis can be automatically performed on multiple data monitored by the monitor to avoid In the process of performing fusion analysis on the monitored multiple data while monitoring the physiological data of different patients at the same time, it is impossible to distinguish whether the monitored data is inaccurate from the same patient. Defects, which improve the accuracy and credibility of the results of data analysis.
  • FIG. 1 is a schematic flow chart of a method for identifying a homology of a physiological signal in an embodiment
  • FIG. 2 is a schematic diagram of waveform matching of two different kinds of physiological signals in one embodiment
  • FIG. 3 is a schematic flow chart of waveform matching in an embodiment
  • FIG. 4 is a schematic diagram showing waveform matching of two different kinds of physiological signals homologous in one embodiment
  • Figure 5 is a schematic illustration of waveform matching of two different kinds of physiological signals homologous in one embodiment
  • FIG. 6 is a schematic diagram of waveform matching of two different kinds of physiological signals of different sources in one embodiment
  • FIG. 7 is a flow chart showing single signal analysis for physiological signals in one embodiment
  • FIG. 8 is a schematic diagram showing the composition between modules of a single signal for physiological signals in one embodiment
  • FIG. 9 is a schematic flow chart of buffer filling using a physiological signal in one embodiment
  • 10 is a flow chart showing whether the signal data of two different kinds of physiological signals are homologously recognized in one embodiment
  • 11 is a schematic diagram showing the structure of each module for identifying whether the signal data of two different kinds of physiological signals are homologous in one embodiment
  • FIG. 12 is a schematic flow chart showing whether the ECG signal and the blood oxygen signal are homologous in one embodiment
  • Figure 13 is a schematic illustration of the homology analysis process of two different kinds of physiological signals in one embodiment
  • FIG. 14 is a schematic structural diagram of a homology identification device for physiological signals in an embodiment
  • Figure 15 is a block diagram showing the structure of a computer device for operating the homology identification method of the aforementioned physiological signal in an embodiment.
  • a method for identifying a homology of a physiological signal is proposed, and the implementation of the method may depend on a computer program, and the computer program may run on a computer system based on a von Neumann system, the computer program may It is an application for homology analysis of physiological signals, and it can also be an application for data homology analysis based on monitor data analysis.
  • the computer system can be a terminal device such as a monitor that runs the computer program described above.
  • the same monitor can monitor related data of a plurality of different types of physiological signals, for example, can simultaneously monitor a patient's ECG signal, blood pressure signal, blood oxygen signal, and the like, and can continuously measure physiological signals.
  • the physiological signal is limited to a physiological signal that can be continuously monitored, because only the data corresponding to the continuous signal can perform historical trend and analysis of the condition monitoring. That is to say, the monitor continuously monitors the patient's various kinds of physiological signals and obtains corresponding signal data for disease analysis.
  • a monitor may also be used to monitor the physiological signals of a number of different patients. In this case, it is necessary to distinguish each of the monitored physiological signals. Whether it belongs to the same patient, so as not to fuse the physiological signals of many different patients. The combined analysis yielded untrustworthy results.
  • the method for identifying the homology of the physiological signal includes the following steps S102-S114:
  • Step S102 Receive signal data of two different kinds of physiological signals.
  • the signal data of different kinds of physiological signals are analyzed for homology
  • whether the signal data of two different kinds of physiological signals are homologously analyzed is analyzed.
  • homology analysis may be performed separately between the plurality of signal data, or homology analysis may be performed for all the monitored signal data.
  • Different types of physiological signals refer to different types or types of physiological signals, at least the measurement principle is different; for example, one is an electrocardiogram signal, and the other is a blood oxygen signal, which is two different kinds of physiological signals.
  • physiological signal A and physiological signal B are two different physiological signals monitored by the monitor.
  • the monitor detects the data analysis module sent by the patient to the monitor, or sends it to other terminal devices specially used for data analysis
  • the physiological signal after the physiological signal, the physiological signal corresponding to the detected physiological signal is acquired.
  • the data concurrently, determines whether the signal data of the different types of physiological signals monitored are homologous.
  • Step S104 Perform waveform matching on signal data of the two different kinds of physiological signals, and determine waveform matching information in the signal data of the two physiological signals.
  • waveform matching of signal data of two different kinds of physiological signals may be performed by matching waveforms in signal data of two different kinds of physiological signals.
  • the waveform corresponding to the signal data of one of the physiological signals is on the upper side
  • the waveform corresponding to the signal data of the other physiological signal is displayed on the lower side, thereby determining the waveform between the two. Matching method.
  • the peak of the physiological signal A corresponds to a peak of the physiological signal B, and in the case of being affected by other external factors, the peak of the physiological signal A and the peak of the physiological signal B cannot be in one-to-one correspondence.
  • waveform matching it can be determined which peak of the physiological signal A corresponds to which peak of the physiological signal B, thereby determining the correspondence between the waveforms in the signal data of the two different kinds of physiological signals.
  • Figure 2 shows An example of a waveform matching, where one is an electrocardiographic signal (EGG, an electrophysiological signal) and the other is a blood oxygen signal (SPO2, a mechanical physiological signal).
  • ECG electrocardiographic signal
  • SPO2 blood oxygen signal
  • the monitor monitors both the ECG signal and the blood oxygen signal
  • the monitored ECG signal and the blood oxygen signal can be simultaneously displayed in the data display window of the monitor.
  • the correspondence between the ECG signal and the peak of the blood oxygen signal is given, that is, the matching method of the peak in the signal data corresponding to the electrocardiogram signal and the blood oxygen signal is determined, and the ECG is determined.
  • Each peak of the signal should correspond to which peak in the signal data of the blood oxygen signal.
  • the waveform matching before the waveform matching is performed, it is also required to determine whether the waveform in the signal data of the acquired physiological signal is valid, for example, in the case of temporarily receiving external interference, the monitoring may be caused.
  • the partial waveform in the signal data of the physiological signal cannot be matched with the waveform of the signal data of the other physiological signal, that is, the partial data in the signal data of the detected physiological signal is invalid.
  • the waveform matching information when determining the waveform matching information, it is necessary to determine whether the time interval between the time stamps corresponding to the corresponding peaks in the signal data corresponding to the physiological signal A and the physiological signal B is at a preset time. Inside, if so, the two peaks are matched. Before determining the waveform matching information, it is also necessary to determine whether the signal data of the monitored physiological signal is valid. For example, when the monitoring is affected by the external signal, the monitored signal data has a great error, and the corresponding Invalid signal data.
  • determining whether the signal data of the detected physiological signal is valid may have multiple determination manners, for example, calculating a signal quality corresponding to the signal type for the signal type of the physiological signal.
  • the parameter; or the signal data that determines whether the signal data of the physiological signal includes the preset invalid condition or the like is not limited in this embodiment.
  • Step S106 Calculate a joint feature between the signal data of the two different kinds of physiological signals according to the waveform matching information.
  • the joint feature refers to a characteristic of a physiological relationship between the physiological signal A and the physiological signal B.
  • the waveform corresponding to the signal data of the electrocardiogram signal and the blood oxygen signal is related to the patient being monitored. Heart rate and other conditions are related.
  • the acquisition of the joint feature may be through the waveform matching information of the signal data of the physiological signal and combined with the other signal characteristics of the physiological signal A and the physiological signal B (for example, the time interval of the time stamp corresponding to the matched peak, The interval of the peaks in the physiological signal, etc.) is used to obtain a characteristic of a physiological relationship between the physiological signal A and the physiological signal B.
  • the joint feature can clearly indicate whether multiple different types of physiological signals are homologous.
  • the step of calculating the joint feature between the signal data of the two different kinds of physiological signals according to the waveform matching information is specifically: determining the two different types according to the waveform matching information. a matching peak in the physiological signal of the kind, calculating a difference sequence of time points corresponding to the matched peaks of the two different kinds of physiological signals, and using the difference sequence as a signal of the two different kinds of physiological signals
  • the joint feature between the data That is to say, the union feature is a sequence of differences in time points corresponding to the matching peaks of the two different kinds of physiological signals.
  • the joint feature includes difference data between the physiological signal A and the time stamp corresponding to the matching peak in the physiological signal B.
  • the peaks in the signal data of the physiological signal B matched by each of the peaks included in the signal data of the physiological signal A are determined, which are the matching peaks.
  • matching peaks in the signal data of the two physiological signals are determined.
  • the time difference of the corresponding time stamps in all the matched peaks is calculated, thereby obtaining a difference sequence.
  • the difference sequence may indicate whether the correspondence between the physiological signal A and the physiological signal B has changed.
  • the physiological signal A is an electrocardiographic signal and the physiological signal B is a blood oxygen signal
  • the peaks or troughs corresponding to the two are related to the heartbeat of the patient being monitored, and therefore, the calculated difference
  • the sequence should be an average sequence (i.e., each element in the sequence is equal), that is, in this embodiment, the difference sequence should be a sequence with a small amplitude.
  • the joint feature may further include a difference sequence of time points corresponding to the matched troughs of the two different kinds of physiological signals, or signal data of two different kinds of physiological signals. a sequence of period lengths between each adjacent crest or trough.
  • Step S108 Calculate a homologous reference coefficient corresponding to the two different kinds of physiological signals according to the joint feature.
  • the homologous reference coefficient may indicate the probability of homology between the two different kinds of physiological signals received in step S102.
  • the value of the homologous reference coefficient may be 0 to 100, and the larger the homologous reference coefficient, the greater the possibility that two different kinds of physiological signals are homologous.
  • calculating the homologous reference coefficients corresponding to the two different kinds of physiological signals according to the joint feature further includes: calculating a formula according to a preset homologous reference coefficient, and the difference sequence, A homologous reference coefficient corresponding to the two different kinds of physiological signals is calculated.
  • homologous reference coefficients for example, calculating an average value, or a variance, a mean square error, and a minimum residual of the difference sequence, and using the calculated result as a homologous reference coefficient corresponding to the two different kinds of physiological signals.
  • the calculation method of the homologous reference coefficient is not limited to the average value of the difference sequence, or the variance, the mean square error, the minimum residual, and the like, and other methods may directly represent the two physiological functions. Whether the signal data of the signal is homologous.
  • Step S110 determining whether the homologous reference coefficient is greater than a preset value, and if yes, performing step S112: determining that the two different kinds of physiological signals are homologous; if not, performing step S114: determining the two different Different types of physiological signals come from different sources.
  • the homologous reference coefficient is a numerical value, and the larger the numerical value, the greater the possibility that the signal data of two different kinds of physiological signals are homologous. On the contrary, the smaller the numerical value, the signal data of two different kinds of physiological signals comes from The less likely it is for the same patient. For example, when the homologous reference coefficient is 100, it is determined that the two different kinds of physiological signals are homologous, and when the homologous reference coefficient is 0, it is determined that the two different kinds of physiological signals are different sources.
  • the threshold of the homologous reference coefficient can be set to 80, and the two different kinds of physiological signals are determined to be homologous when the calculated homologous reference coefficient is greater than 80. Otherwise, the two different kinds of physiological signals are determined to be different. source.
  • Fig. 4 and Fig. 5 a schematic diagram of waveform matching of two different kinds of physiological signals of homology is given.
  • each peak corresponds to each other. And the corresponding relationship does not change; in FIG. 5, among the effective signal data of the two physiological signals, each peak corresponds to each other, and the correspondence does not change.
  • the signal data between the two physiological signals is not Each peak has a one-to-one correspondence, and its correspondence also has an abnormality.
  • the two physiological signals shown in Fig. 6 are different sources.
  • Steps S102-S114 are to determine whether the two different kinds of physiological signals monitored are homologous by calculating the homologous reference coefficients of the two different kinds of physiological signals, and step S102 is a step of signal acquisition, and steps S104-S114 are The process of signal homology analysis.
  • the optional step further includes pre-processing the received signal data.
  • the preprocessing method includes filtering the signal data.
  • the method further includes: performing high-pass filtering processing and low-pass filtering processing on the signal data of the two different kinds of physiological signals.
  • the high-pass filtering process on the signal data can be implemented by a high-pass filter. After high-pass filtering processing on the signal data, the low-frequency signal below the preset threshold is blocked and attenuated, that is, the data lower than the preset frequency will be Attenuated, it can be used to eliminate low frequency noise in the signal data of physiological signals.
  • the low-boiling filter processing of the signal data can be implemented by a preset low-pass filter. After low-pass filtering for the monitored signal data, the high-frequency signal higher than the preset threshold value is blocked and attenuated. That is to say, data higher than the preset frequency is attenuated and can be used to eliminate high frequency noise in the signal data of the monitored physiological signal.
  • a high pass filtering process (cutoff frequency of 0.05 Hz) and a low pass filtering process (cutoff frequency of 40 Hz) are performed for the electrocardiographic signal (EGG).
  • a high pass filtering process (cutoff frequency of 0.3 Hz) and a low pass filtering process (cutoff frequency of 5 Hz) are performed for the blood oxygen signal (SPO2).
  • the acquisition and analysis of the signal data of the above physiological signals are based on the analysis of physiological signals based on temporal changes or signal characteristics, and are equivalent to extracting the characteristics of the signal data of the physiological signals in the time domain.
  • it may also be based on features of the physiological signal in the frequency domain.
  • the method before performing waveform matching on the signal data of the two different kinds of physiological signals, the method further includes: performing signal data on the signal data of the two different kinds of physiological signals separately Transforming the data to obtain the transformed signal data; performing waveform matching on the signal data of the two different kinds of physiological signals: performing waveform matching on the transformed signal data corresponding to the signal data of the two different kinds of physiological signals .
  • the transformation used may be not only a Fourier transform, but also any other A transform that transforms time domain data into frequency domain data can be implemented.
  • any transformation such as wavelet transform or cosine transform, may be used to change the signal data of the originally received physiological signal and then perform processing.
  • the signal data of the received physiological signal is subjected to Fourier transform or the like, the signal data of the physiological signal may be received, or after the signal data is preprocessed, but it is required Before waveform matching of signal data.
  • the corresponding feature extraction algorithm may be extracted according to a preset feature extraction algorithm, and then performed on the feature data. analysis.
  • the method further includes: extracting feature data of the signal data of the two different kinds of physiological signals, respectively.
  • different types of physiological signals have different characteristic data.
  • the process of extracting the feature data corresponding to the signal data may be a process of detecting and classifying the QRS complex (QRS complex) of the signal data, wherein the QRS wave reflects the heart at the time of ventricular contraction
  • QRS complex QRS complex
  • the electrical behavior is the basis for the automatic analysis of ECG signals.
  • the process of extracting the feature data corresponding to the signal data is a process of monitoring the PLUS wave (pulse wave) corresponding to the signal data of the SPO2, and the PLUS wave is the characteristic data reflecting the blood oxygen signal.
  • the feature data of the signal data of the physiological signal is extracted, the feature data can be performed according to the characteristic data.
  • the feature analysis for example, the feature data corresponding to the signal data of the physiological signal is used as the signal data of the physiological signal monitored in step S102 to perform the homology analysis in steps S104-S114.
  • corresponding data analysis is performed on the feature data corresponding to the signal data of the single physiological signal, for example, calculating a signal quality parameter corresponding to the signal data of the physiological signal to determine the detected signal data. Whether the preset quality standard is reached, if it is reached, the monitored signal data is up to standard, and further data analysis or other operations can be performed. Otherwise, if it is not reached, the monitored signal data is not up to standard, and if it is used for further The data analysis results in the data analysis results may be due to the quality of the signal data is not up to standard, resulting in large errors, that is, the reliability of the data analysis results is insufficient.
  • the signal quality index can reflect the signal quality corresponding to the signal data of the monitored physiological signal, and the signal quality parameter ranges from 0 to 100.
  • the higher the signal quality parameter the corresponding The higher the quality of the signal data of the physiological signal.
  • the signal quality parameter can give an objective assessment of the quality of the monitored signal data of the physiological signal.
  • the physiological signal is an electrocardiographic signal
  • the corresponding characteristic data is a QRS wave
  • the signal-to-noise ratio (SNR) of the QRS wave may be calculated first. Then, the signal quality parameter corresponding to the ECG signal is calculated by the signal to noise ratio, or the signal to noise ratio is directly used as the signal quality parameter corresponding to the ECG signal.
  • the signal quality parameter is calculated to determine whether the currently monitored physiological signal satisfies the condition. Specifically, whether the signal quality parameter corresponding to the monitored physiological signal signal data satisfies a preset signal quality parameter threshold. If it is satisfied, it is determined that it satisfies the preset condition, and further processing is possible. On the other hand, if the signal quality parameter corresponding to the monitored physiological signal signal data does not satisfy the preset signal quality parameter threshold, the signal data of the corresponding physiological signal is not If the preset conditions are met, if it is used for further data analysis, a large error may occur, and therefore, the signal data of the physiological signal is considered to be invalid.
  • the signal data of the monitored physiological signal it is not only necessary to determine whether it satisfies a certain signal quality, but also whether it is abnormal or otherwise.
  • the signal data corresponding to the ECG signal it is also necessary to determine whether there is a heart rate abnormality. If there is a heart rate abnormality, it is necessary to directly give an alarm prompt.
  • the method further includes: determining, for any kind of physiological signals, a signal type of the physiological signal, and determining a single signal parameter corresponding to the signal type.
  • Type calculating a single signal parameter corresponding to the physiological signal according to a preset single signal parameter calculation formula and the characteristic data; determining whether the single signal parameter satisfies a preset single signal parameter threshold; and the single signal parameter
  • the preset single-signal parameter threshold is not met, a message indicating that the signal is abnormal is generated and the user is prompted.
  • the corresponding signal abnormality or the need to prompt the situation will be different.
  • the situation that needs to be prompted is different from the abnormal heart rate of the ECG signal, but Prompt when hypoxemia is too low. Therefore, it is first necessary to determine a single signal parameter type corresponding to the signal type according to the signal type of the physiological signal, for example, the ECG signal is the calculated heart rate, and the blood oxygen signal is the calculated blood oxygen concentration.
  • the single signal parameter corresponding to the signal parameter of the monitored physiological signal can be calculated according to the preset single signal parameter calculation formula. It should be noted that, in this embodiment, when calculating a single signal parameter, the calculation may be directly performed by the signal data of the physiological signal, or may be calculated by using the feature data corresponding to the signal data of the extracted physiological signal. of.
  • the signal type of the physiological signal is an electrocardiogram signal
  • the corresponding single signal parameter type is a heart rate
  • the position information of the QRS wave in the QRS wave data extracted from the signal data of the electrocardiographic signal is calculated. Heart rate.
  • the signal type of the physiological signal is a blood oxygen signal
  • the corresponding single signal parameter type is blood oxygen concentration
  • the position information of the PLUS wave in the PLUS wave data extracted from the signal data of the electrocardiographic signal is The corresponding pulse is calculated, and the blood oxygen concentration corresponding to the blood oxygen signal is calculated according to the AC and DC component ratio of the PLUS wave.
  • the single signal parameter meter corresponding to the signal data of the physiological signal
  • it is also necessary to determine whether the single signal parameter satisfies a preset condition for example, determining whether the single signal parameter satisfies a preset single signal parameter threshold. For example, if the heart rate exceeds a certain value or the heart rate is lower than a certain value, it is an abnormal situation.
  • a single signal parameter threshold corresponding to the corresponding single signal parameter type is set to determine whether the calculated single signal parameter satisfies the single signal parameter threshold.
  • the next step of the homology analysis can be continued; if not, the currently monitored physiological signal is indicated.
  • the user needs to be prompted for the user to process in time. For example, in the case of a low heart rate, the user is prompted in the form of an alarm.
  • the single signal parameter when the single signal parameter does not meet the preset single signal parameter threshold, not only the corresponding prompt information needs to be generated to prompt the user, but also whether to continue the homology according to the actual situation.
  • sexual analysis Because, in the case of an abnormality in the signal data of the physiological signal, one of the signal data is abnormal, and other signal data may be abnormal or may not be abnormal. In this case, two different physiology are also required. Whether the signals are homologous for analysis to improve the reliability of the patient's condition analysis.
  • a data analysis process for a single physiological signal of a plurality of steps such as quality parameters, calculation of single signal parameters, and the like.
  • the process of performing data preprocessing, feature data extraction, calculating signal quality parameters, and calculating single signal parameters on the signal data of the received single physiological signal is a process of single signal analysis.
  • Figure 7 illustrates the process of single signal analysis for physiological signal A.
  • the signal data is first preprocessed to improve the reliability of the subsequent data analysis; then the feature data of the signal data is extracted, and the corresponding signal quality parameters are respectively calculated.
  • a single signal parameter and judge whether the calculated signal quality parameter and the single signal parameter satisfy the preset condition, and if both of the preset conditions are met, proceed to the next step, for example, performing waveform matching and determining waveform matching information .
  • each step or operation of single signal analysis is performed by a corresponding module.
  • the process of preprocessing the signal data of the physiological signal A is performed by the data.
  • Pre-processing module to complete.
  • Figure 8 shows the data relationship between the various modules during the single signal analysis process.
  • the signal data of the physiological signal of the patient monitored by the monitor analyzes the physical condition of the patient, for example, analyzing the patient's development trend.
  • the relevant data of the patient within a certain period of time is required for accurate analysis.
  • the data to be analyzed is only the monitored physiological signal within 30s of the patient.
  • the monitored signal data can only indicate the patient's condition within 30s, and may be temporarily affected within 30s, and cannot fully represent the patient's condition. Therefore, in general, in order to improve the accuracy of the analysis result of analyzing the signal data of the physiological signal monitored by the monitor, in the embodiment, the signal data monitored by the monitor needs to reach a certain amount of data.
  • the method before performing waveform matching on the signal data of the two different kinds of physiological signals, the method further includes: determining, according to the signal data of the two different kinds of physiological signals, whether the data size of the signal data is greater than or equal to a preset data amount threshold, if yes, performing waveform matching on the signal data of the two different kinds of physiological signals; if not, continuing to perform the signal data receiving the two different kinds of physiological signals.
  • a certain data amount threshold is set, and the monitored data is analyzed only when the amount of data corresponding to the monitored signal data reaches or exceeds the data amount threshold. That is to say, only when the amount of data corresponding to the monitored signal data reaches or exceeds the preset data amount threshold, the homology analysis of the monitored signal data of different kinds of physiological signals is performed. Otherwise, the monitoring of the signal data of the physiological signal is continued to achieve the preset amount of data.
  • the signal data of the monitored physiological signal may be buffer-filled, and the storage amount of the buffer is set in advance. Then, the homology analysis is performed only after the signal data of the monitored physiological signal fills the buffer and accumulates to a certain amount of data, for example, the signal data of the detected physiological signal overflows the buffer. In the case of waveform matching or data preprocessing of different kinds of physiological signals detected.
  • step S104 the next operation step is continued, for example, step S104 is performed; otherwise, the monitoring of the signal data of the physiological signal is returned, that is, the initial step of returning to the homology analysis.
  • FIG. 10 shows a flow diagram of an identification process for whether the physiological signal A and the physiological signal B are homologous.
  • the signal data of the monitored physiological signal A and the signal data of the physiological signal B are first subjected to single signal analysis (ie, signal preprocessing, characteristics). Data extraction, signal quality parameter calculation, single signal parameter calculation, etc., and then waveform matching is performed to determine waveform matching information between signal signals of two physiological signals, and corresponding joint characteristics are determined according to waveform matching information, and physiological parameters are calculated.
  • single signal analysis ie, signal preprocessing, characteristics
  • waveform matching is performed to determine waveform matching information between signal signals of two physiological signals, and corresponding joint characteristics are determined according to waveform matching information, and physiological parameters are calculated.
  • the homologous reference parameter between the signal A and the physiological signal B finally determines whether the physiological signal A and the physiological signal B are homologous according to the homologous reference parameter. Further, the composition of the module in the process of identifying whether the physiological signal A and the physiological signal B are homologous, that is, the signal monitoring module, the single signal analysis module, and the signal homology analysis module are given in FIG.
  • the physiological signal A is an electrocardiographic signal (EGG)
  • the physiological signal B is a blood oxygen signal (SPO2)
  • the ECG signal detected by the monitor and the blood oxygen signal are homologous.
  • the process can be as shown in Figure 12.
  • the physiological signal in addition to the feature data, the signal quality parameter, the single signal singular number, the joint feature, and the homologous reference coefficient corresponding to the signal data of the physiological signal, the physiological signal can be acquired.
  • the difference between the A and the physiological signal B (the difference between the lengths of the two waveform change periods under the same physiological signal) and the mean peak-to-peak difference (the time between the peak of the physiological signal A and the peak of the physiological signal B)
  • the maximum value the maximum value of the difference between the peak of the physiological signal A and the timestamp corresponding to the peak of the physiological signal B
  • the minimum value the peak of the physiological signal A and the physiological signal B
  • the obtained parameters are not limited to the above enumerated parameters, and may also include other data that can be displayed between two signals. Other parameters or characterization data of homology relationships.
  • the ratio of the synchronization between the peak data of the physiological signal A and the peak data of the physiological signal B is calculated, and the historical data based on the ratio and the homology are calculated. It is determined whether the physiological signal A and the physiological signal B are homologous.
  • a physiological signal homology identification device including a signal data receiving module 102, a waveform matching module 104, a joint feature acquiring module 106, and a homologous reference coefficient calculating module. 108.
  • the homology identification module 110 wherein:
  • the signal data receiving module 102 is configured to receive signal data of two different kinds of physiological signals
  • the waveform matching module 104 is configured to perform waveform matching on signal data of the two different kinds of physiological signals, and determine waveform matching information in signal data of the two physiological signals;
  • the joint feature acquiring module 106 is configured to calculate a joint feature between the signal data of the two different kinds of physiological signals according to the waveform matching information;
  • the homologous reference coefficient calculation module 108 is configured to calculate a homologous reference coefficient corresponding to the two different kinds of physiological signals according to the joint feature;
  • the homology identification module 110 is configured to determine that the two different kinds of physiological signals are homologous if the homologous reference coefficient is greater than a preset value.
  • the joint feature acquiring module 106 is further configured to determine, according to the waveform matching information, matching peaks in the two different kinds of physiological signals, and calculate the two different kinds of physiological signals. A sequence of differences of time points corresponding to the matched peaks, the difference sequence being used as a joint feature between the signal data of the two different kinds of physiological signals.
  • the foregoing apparatus further includes a data pre-processing module 112, configured to perform high-pass filtering processing and low-pass filtering on signal data of the two different kinds of physiological signals. deal with.
  • the apparatus further includes a signal quality parameter calculation module 114, configured to separately extract feature data of signal data of the two different kinds of physiological signals; Setting a signal quality parameter calculation formula to calculate a signal quality parameter corresponding to the feature data; and if the signal quality parameter does not satisfy a preset signal quality parameter threshold, determining the physiological signal corresponding to the signal quality parameter The signal data is invalid; the waveform matching module 104 is invoked if the signal quality parameter satisfies a preset signal quality parameter threshold.
  • a signal quality parameter calculation module 114 configured to separately extract feature data of signal data of the two different kinds of physiological signals. Setting a signal quality parameter calculation formula to calculate a signal quality parameter corresponding to the feature data; and if the signal quality parameter does not satisfy a preset signal quality parameter threshold, determining the physiological signal corresponding to the signal quality parameter The signal data is invalid; the waveform matching module 104 is invoked if the signal quality parameter satisfies a preset signal quality parameter threshold.
  • the apparatus further includes a single signal parameter calculation module 116, configured to determine a signal type of the physiological signal for any kind of physiological signal, and determine the signal a single signal parameter type corresponding to the type; calculating a single signal parameter corresponding to the physiological signal according to the preset single signal parameter calculation formula and the characteristic data; determining whether the single signal parameter satisfies a preset single signal parameter threshold; Performing waveform matching on signal data of the two different kinds of physiological signals when the single signal parameter satisfies a preset single signal parameter threshold; if the single signal parameter does not satisfy a preset single signal parameter At the threshold, a message indicating that the signal is abnormal is generated and the user is prompted.
  • a single signal parameter calculation module 116 configured to determine a signal type of the physiological signal for any kind of physiological signal, and determine the signal a single signal parameter type corresponding to the type; calculating a single signal parameter corresponding to the physiological signal according to the preset single signal parameter calculation formula and the characteristic data; determining whether the single signal parameter satis
  • the apparatus further includes a buffer filling module 118, configured to determine data of the signal data for signal data of the two different kinds of physiological signals. Whether the size is greater than or equal to a preset data amount threshold, and in a case where the data amount of the signal data is greater than or equal to a preset data amount threshold, the waveform matching module 104 is invoked; data in the signal data The signal data receiving module 102 is invoked if the amount of size is less than a preset data amount threshold.
  • a buffer filling module 118 configured to determine data of the signal data for signal data of the two different kinds of physiological signals. Whether the size is greater than or equal to a preset data amount threshold, and in a case where the data amount of the signal data is greater than or equal to a preset data amount threshold, the waveform matching module 104 is invoked; data in the signal data The signal data receiving module 102 is invoked if the amount of size is less than a preset data amount threshold.
  • the homology reference coefficient calculation module 108 is further configured to calculate a formula corresponding to the two different kinds of physiological signals according to a preset homology reference coefficient calculation formula and the difference sequence. Homology reference coefficient.
  • the homology reference coefficient calculation module 108 is further configured to calculate an average value or a mean square error of the difference sequence as a homologous reference coefficient corresponding to the two different kinds of physiological signals. .
  • the apparatus further includes a signal data transformation module 120, configured to separately perform Fourier transform on signal data of the two different kinds of physiological signals, to obtain Transforming the signal data; the waveform matching module is further configured to perform waveform matching on the transformed signal data corresponding to the signal data of the two different kinds of physiological signals.
  • a signal data transformation module 120 configured to separately perform Fourier transform on signal data of the two different kinds of physiological signals, to obtain Transforming the signal data
  • the waveform matching module is further configured to perform waveform matching on the transformed signal data corresponding to the signal data of the two different kinds of physiological signals.
  • the monitor monitors signal data of a plurality of different kinds of physiological signals
  • the joint characteristics of the signal data of the monitored different kinds of physiological signals are adopted.
  • the extraction and calculations determine the size of the homologous reference parameter that can identify the likelihood that two different kinds of physiological signals are homologous, thereby determining whether the two physiological signals are homologous. That is to say, the homology analysis can be automatically performed on a plurality of data monitored by the monitor, and the monitor can avoid the fusion analysis of the monitored plurality of data while monitoring the physiological data of different patients at the same time. Because it is impossible to distinguish whether the monitored data is a defect from the same patient resulting in inaccurate analysis results, thereby improving the accuracy and credibility of the results of the data analysis.
  • the above embodiments it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • a software program it may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions.
  • the computer program instructions When the computer program instructions are loaded and executed on a computer, the processes or functions described in accordance with embodiments of the present invention are generated in whole or in part.
  • the computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable device.
  • the computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be from a website site, computer, server or data center Transfer to another website site, computer, server, or data center by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL), or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer readable storage medium can be any available media that can be accessed by a computer or a data storage device such as a server, data center, or the like that includes one or more available media.
  • the usable medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (such as a solid state disk (SSD)) or the like.
  • a magnetic medium eg, a floppy disk, a hard disk, a magnetic tape
  • an optical medium eg, a DVD
  • a semiconductor medium such as a solid state disk (SSD)
  • FIG. 15 illustrates a terminal of a von Neumann system-based computer system that operates the homology identification method of the physiological signals described above.
  • the computer system can be a terminal device such as a smartphone, a tablet, a palmtop, a laptop, or a personal computer.
  • an external input interface 1001, a processor 1002, a memory 1003, and an output interface 1004 connected through a system bus may be included.
  • the external input interface 1001 can optionally include at least a network interface 10012.
  • the memory 1003 may include an external memory 10032 (eg, a hard disk, an optical disk, or a floppy disk, etc.) and an internal memory 10034.
  • the output interface 1004 can include at least a device such as a display 10042.
  • the operation of the method is based on a computer program whose program file is stored in the external memory 10032 of the aforementioned von Neumann system-based computer system, loaded into the internal memory 10034 at runtime, and then After being compiled into machine code, it is passed to the processor 1002 for execution, so that the logical signal data receiving module 102, the waveform matching module 104, the joint feature acquiring module 106, and the homologous reference are formed in the von Neumann system-based computer system.
  • the input parameters are all received through the external input interface 1001, and transferred to the buffer in the memory 1003, and then input to the processor 1002 for processing, the processed result data or the cache. Subsequent processing is performed in the memory 1003 or passed to the output interface 1004 for output.
  • processor 1002 is configured to perform the following operations:
  • the two different kinds of physiological signals are determined to be homologous.
  • the processor 1002 is further configured to determine, according to the waveform matching information, matching peaks in the two different kinds of physiological signals, and calculate the two different kinds of physiological A sequence of differences of time points corresponding to the matched peaks in the signal, the difference sequence being used as a joint feature between the signal data of the two different kinds of physiological signals.
  • the processor 1002 is further configured to perform high-pass filtering processing and low-pass filtering processing on signal data of the two different kinds of physiological signals.
  • the processor 1002 is further configured to perform feature data for extracting signal data of the two different kinds of physiological signals, respectively; calculating a formula and calculating the feature data according to a preset signal quality parameter. Corresponding signal quality parameter; if the signal quality parameter does not satisfy the preset signal quality parameter threshold, determining that the signal data of the physiological signal corresponding to the signal quality parameter is invalid; in the signal quality parameter In the case that the preset signal quality parameter threshold is satisfied, the waveform matching of the signal data of the two different kinds of physiological signals is performed.
  • the processor 1002 is further configured to: determine, for any kind of physiological signal, a signal type of the physiological signal, determine a single signal parameter type corresponding to the signal type; according to a preset Calculating a signal parameter formula and the characteristic data, calculating a single signal parameter corresponding to the physiological signal; determining whether the single signal parameter satisfies a preset single signal parameter threshold; and the single signal parameter does not satisfy a preset single signal
  • the parameter threshold is used, a message indicating that the signal is abnormal is generated and the user is prompted.
  • the processor 1002 is further configured to perform signal data for the two different kinds of physiological signals, and determine whether the data size of the signal data is greater than or equal to a preset data amount threshold. And if yes, performing waveform matching on the signal data of the two different kinds of physiological signals; if not, continuing to perform the signal data receiving the two different kinds of physiological signals.
  • the processor 1002 is further configured to perform a homology reference corresponding to the two different kinds of physiological signals according to a preset homology reference coefficient calculation formula and the difference sequence. coefficient.
  • the processor 1002 is further configured to perform calculating an average value or a mean square error of the difference sequence as a homologous reference coefficient corresponding to the two different kinds of physiological signals.
  • the processor 1002 is further configured to separately perform Fourier transform on the signal data of the two different kinds of physiological signals to obtain transformed signal data; and the two different kinds of physiological The converted signal data corresponding to the signal data of the signal is waveform matched.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Public Health (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Physiology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Mathematical Physics (AREA)
  • Cardiology (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Pulmonology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

一种生理信号的同源性识别方法及装置，其中该方法包括：接收两种不同种类的生理信号的信号数据（S102）；对该两种不同种类的生理信号的信号数据进行波形匹配，确定该两种生理信号的信号数据中的波形匹配信息（S104）；根据该波形匹配信息计算该两种不同种类的生理信号的信号数据之间的联合特征（S106）；根据该联合特征计算与该两种不同种类的生理信号对应的同源参考系数（S108）；在该同源参考系数大于预设值的情况下，确定该两种不同种类的生理信号同源。采用该生理信号的同源性识别方法及装置，可判断监护仪中的多种信号是否来自于同一个病人，提高监护仪的数据分析的可信度。

Description

生理信号的同源性识别方法及装置 技术领域
本发明涉及医学仪器控制的技术领域，尤其涉及一种生理信号的同源性识别方法及装置。
背景技术
随着医学技术的发展以及终端技术的发展，对病人的各种生理状况进行检测的监护仪的功能也从单一的生理信号的监测发展到了多种生理信号的监测，对病人的生理信号进行分析的方法也从单一信号的分析发展到了多参数融合分析，以提高分析结果的可靠性和准确度。即，同一监护仪可以对病人的多种生理信号进行监测，例如，可以对病人的心电信号、血压、血氧信号等多项参数同时进行检测，并根据获取到的多项参数的观测数据对用户当前的身体状态进行分析和判断，相较于对用户当前的单一生理信号的观测结果进行分析的方案，提高了分析结果的可靠性和准确度。
但是，虽说监护仪可以同时监测病人的多项生理参数并进行融合分析，减少了监护仪的误报或漏报，但是，在监护仪短缺或者其他特殊情况下，也可能使用一台监护仪来监测两个或者多个病人的不同种类的生理信号，例如，使用监护仪A监测病人B的心电信号，并同时监测病人C的血压以及血氧信号。而对于现有的监护仪来说，并不能区别当前检测到的多种信号是否来自于同一个病人，在进行多参数融合分析时，只会根据当前仪器检测到的所有的信号进行分析，而不会去区分不同的信号所来自的病人。也就是说，一旦监护仪被用于对不止一个病人的生理信号的监测，监护仪在进行融合分析时会根据多个病人的不同种类的生理信号进行分析，因为进行融合分析的多种心里信号来自于不同的病人，并且各个病人的具体情况存在差异，这就会产生分析的结果出现严重的误报或者漏报，导致多参数融合分析的可信度大幅下降。
发明内容
基于此，为解决传统技术中的对监护仪监测到的多种生理信号进行多参数 融合分析时因为不能区别当前的多种生理信号是否来自于同一个病人而存在的可信度不足的技术问题，特提出了一种生理信号的同源性识别方法。
一种生理信号的同源性识别方法，包括：
接收两种不同种类的生理信号的信号数据；
对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
可选的，在其中一个实施例中，所述根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征包括：
根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。可选的，在其中一个实施例中，所述接收两种不同种类的生理信号的信号数据之后还包括：
对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
可选的，在其中一个实施例中，所述接收两种不同种类的生理信号的信号数据之后还包括：
分别提取所述两种不同种类的生理信号的信号数据的特征数据；
根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；
在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；
在所述信号质量参数满足预设的信号质量参数阈值的情况下，执行所述对 所述两种不同种类的生理信号的信号数据进行波形匹配。
可选的，在其中一个实施例中，所述分别提取所述两种不同种类的生理信号的信号数据的特征数据之后还包括：
针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；
根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；
判断所述单信号参数是否满足预设的单信号参数阈值；
在所述单信号参数满足预设的单信号参数阈值时，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；
在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
可选的，在其中一个实施例中，所述对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包括：
针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，若是，则执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；若否，则继续执行所述接收两种不同种类的生理信号的信号数据。
可选的，在其中一个实施例中，所述根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数还包括：
根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在其中一个实施例中，所述根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数还包括：
计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在其中一个实施例中，所述对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包括：
对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变 换信号数据；
所述对所述两种不同种类的生理信号的信号数据进行波形匹配具体为：
对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
此外，为解决传统技术中的对监护仪监测到的多种生理信号进行多参数融合分析时因为不能区别当前的多种生理信号是否来自于同一个病人而存在的可信度不足的技术问题，特提出了一种生理信号的同源性识别装置。
一种生理信号的同源性识别装置，包括：
信号数据接收模块，用于接收两种不同种类的生理信号的信号数据；
波形匹配模块，用于对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
联合特征获取模块，用于根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
同源参考系数计算模块，用于根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
同源识别模块，用于在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
可选的，在其中一个实施例中，所述联合特征获取模块还用于根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。
可选的，在其中一个实施例中，所述装置还包括数据预处理模块，用于对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
可选的，在其中一个实施例中，所述装置还包括信号质量参数计算模块，用于分别提取所述两种不同种类的生理信号的信号数据的特征数据；根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；在所述信号质量参数满足预设的信号质量参数阈值的情况下，调用所述波形匹配模块。
可选的，在其中一个实施例中，所述装置还包括单信号参数计算模块，用于针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；判断所述单信号参数是否满足预设的单信号参数阈值；在所述单信号参数满足预设的单信号参数阈值时，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
可选的，在其中一个实施例中，所述装置还包括缓冲区填充模块，用于针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，在所述信号数据的数据量大小大于或等于预设的数据量阈值的情况下，调用所述波形匹配模块；在所述信号数据的数据量大小小于预设的数据量阈值的情况下，调用所述信号数据接收模块。
可选的，在其中一个实施例中，所述同源参考系数计算模块还用于根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在其中一个实施例中，所述同源参考系数计算模块还用于计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在其中一个实施例中，所述装置还包括信号数据变换模块，用于对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；
所述波形匹配模块还用于对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
实施本发明实施例，将具有如下有益效果：
采用了上述生理信号的同源性识别方法和装置之后，在监护仪监测到多种不同种类的生理信号的信号数据的情况下，通过对监测到的不同种类的生理信号的信号数据的联合特征的提取和计算，来确定可以标识两种不同种类的生理信号是同源的可能性的同源参考参数的大小，从而判断两个生理信号是否是同源的。也就是说，可以对监护仪监测到的多项数据自动进行同源性分析，避免 了监护仪在同时监测不同的病人的生理数据的情况下对监测到的多项数据进行融合分析的过程中，因为无法区别监测到的数据是否是来自于同一个病人导致的分析结果不准确的缺陷，从而提高了数据分析的结果的准确度和可信度。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
其中：
图1为一个实施例中一种生理信号的同源性识别方法的流程示意图；
图2为一个实施例中两种不同种类的生理信号的波形匹配示意图；
图3为一个实施例中波形匹配的流程示意图；
图4为一个实施例中同源的两种不同种类的生理信号的波形匹配的示意图；
图5为一个实施例中同源的两种不同种类的生理信号的波形匹配的示意图；
图6为一个实施例中不同源的两种不同种类的生理信号的波形匹配的示意图；
图7为一个实施例中针对生理信号进行单信号分析的流程示意图；
图8为一个实施例中针对生理信号进行单信号分析的各个模块之间的组成示意图；
图9为一个实施例中使用生理信号进行缓冲区填充的流程示意图；
图10为一个实施例中针对两种不同种类的生理信号的信号数据是否同源进行识别的流程示意图；
图11为一个实施例中针对两种不同种类的生理信号的信号数据是否同源进行识别的各个模块之间的结构示意图；
图12为一个实施例中对心电信号和血氧信号是否同源进行识别的流程示意图；
图13为一个实施例中两种不同种类的生理信号的同源性分析过程的示意图；
图14为一个实施例中一种生理信号的同源性识别装置的结构示意图；
图15为一个实施例中运行前述生理信号的同源性识别方法的计算机设备的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
为解决传统技术中的对监护仪监测到的多种生理信号进行多参数融合分析时因为不能区别当前的多种生理信号是否来自于同一个病人而存在的可信度不足的技术问题，在本实施例中，特提出了一种生理信号的同源性识别方法，该方法的实现可依赖于计算机程序，该计算机程序可运行于基于冯诺依曼体系的计算机系统之上，该计算机程序可以是生理信号的同源性分析的应用程序，也可以是基于监护仪的数据分析的数据同源性分析的应用程序。该计算机系统可以是运行上述计算机程序的例如监护仪的终端设备。
在本实施例中，同一个监护仪可以监测多种不同类型的生理信号的相关数据，例如，可以同时监测病人的心电信号、血压信号、血氧信号等多种可以连续进行测量的生理信号。需要说明的是，在本实施例中，生理信号限于可以连续进行监测的生理信号，因为只有连续不断的信号所对应的数据才可以进行历史趋势以及病情监测的分析。也就是说，监护仪连续不断的监测病人的多种不同种类的生理信号，并获取对应的信号数据，以用于病情分析。
在医护人员误操作或者在设备短缺的情况下，一台监护仪也可能被用于监测多个不同的病人的生理信号，在这种情况下，就需要区分每一种监测到的生理信号之间是否是属于同一个病人，以免对多个不同的病人的生理信号进行融 合分析得到不可信的结果。
具体的，如图1所示，上述生理信号的同源性识别方法包括如下步骤S102-S114：
步骤S102：接收两种不同种类的生理信号的信号数据。
在本实施例中，在对不同种类的生理信号的信号数据是否同源进行分析时，是针对两种不同种类的生理信号的信号数据是否同源进行分析的。在监护仪监测到不止两种类型的生理信号的信号数据时，可以分别在多个信号数据之间两两进行同源性分析，或者，针对所有的监测到的信号数据进行同源性分析。
生理信号的不同种类，是指生理信号的类型或者种类不同，至少是测量原理不同；例如，其一为心电信号，其二为血氧信号，为两种不同种类的生理信号。为了方便叙述，将监护仪监测的两种不同同类的生理信号分别称为生理信号A和生理信号B。
在本实施例中，在监护仪检测到病人的送给监护仪中的数据分析模块，或者发送给其他专门用于数据分析的终端设备，以生理信号之后，即获取与检测到的生理信号对应的数据，并发确定监测到的不同种类的生理信号的信号数据是否是同源的。
步骤S104：对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息。
在本实施例中，对两种不同种类的生理信号的信号数据进行波形匹配的方式，可以是针对两种不同种类的生理信号的信号数据中的波形进行匹配。例如，在同一个波形图像展示页面中，其中一个生理信号的信号数据所对应的波形在上，下方对应的展示另一个生理信号的信号数据所对应的波形，从而确定二者之间的波形的匹配方式。
正常情况下，生理信号A的波峰对应生理信号B的的一个波峰，而在受到其他外因的影响的情况下，生理信号A的波峰与生理信号B的波峰将无法一一对应。通过波形匹配，可以确定生理信号A的波峰具体对应了生理信号B的哪一个波峰，从而确定两种不同种类的生理信号的信号数据中的波形之间的对应关系。
以确定波峰之间的匹配关系为例，在图2所示的应用场景中，图2展示了 一个波形匹配的实例图，其中，一为心电信号(EGG，一种电生理信号)，一为血氧信号(SPO2，一种机械生理信号)。在监护仪同时监测心电信号和血氧信号时，可以在监护仪的数据显示窗口中同时展示监测到的心电信号以及血氧信号。在图2中，给出了心电信号与血氧信号的波峰之间的对应关系，即确定了心电信号与血氧信号对应的信号数据中的波峰的匹配方式，并且，确定了心电信号的每一个波峰应该对应血氧信号的信号数据中的哪一个波峰。
可选的，在本实施例中，在进行波形匹配之前，还需要判断获取到的生理信号的信号数据中的波形是否是有效的，例如，在暂时受到外因干扰的情况下，可能导致监测到的生理信号的信号数据中的部分波形无法与另一个生理信号的信号数据的波形进行匹配，即检测到的生理信号的信号数据中的部分数据是无效的。
具体的，在一个可选的实施例中，在对所述两种不同种类的生理信号的信号数据进行波形匹配并确定所述两种生理信号的信号数据中的波形匹配信息之前，还需要确定生理信号A以及生理信号B所分别对应的信号数据中的数据时有效的，即确定信号数据中的波峰或波谷等数据是有效的。
如图3所示，在确定上述波形匹配信息时，需要确定生理信号A以及生理信号B分别对应的信号数据中，对应的波峰所对应的时间戳之间的时间间隔是否在预设的时间之内，若是，则这两个波峰是匹配的。而在确定波形匹配信息之前，还需要判断监测到的生理信号的信号数据是否是有效的，例如，在监测时受到外因信号的影响，导致监测到的信号数据存在极大的误差，则对应的信号数据时无效的。
需要说明的是，在本实施例中，判断检测到的生理信号的信号数据是否是有效的，可以有多种判断方式，例如，针对生理信号的信号类型，计算与该信号类型对应的信号质量参数；或者，判断该生理信号的信号数据是否包含的满足预设的无效条件的信号数据等，在本实施例中并不做限制。
步骤S106：根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征。
联合特征是指生理信号A与生理信号B之间具有典型生理对应关系的特征，例如，心电信号与血氧信号的信号数据所对应的波形均与被监测的病人的 心跳频率等情况相关。在本实施例中，联合特征的获取，可以通过对生理信号的信号数据的波形匹配信息并结合生理信号A与生理信号B的其他信号特征(例如，匹配的波峰对应的时间戳的时间间隔、生理信号中波峰的间隔时间等)来获取生理信号A与生理信号B之间具有典型生理对应关系的特征。一般来讲，联合特征可以清晰的表明多个不同种类的生理信号是否是同源的。
在一个具体的实施例中，上述根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征的步骤具体为：根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。也就是说，联和特征为两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列。
具体的，联合特征包括有生理信号A与生理信号B中匹配的波峰所对应的时间戳之间的差值数据。例如，在步骤S104中，确定了生理信号A的信号数据中所包含的每一个波峰所匹配的生理信号B的信号数据中的波峰，二者即为匹配的波峰。例如，在图2所示，确定了两个生理信号的信号数据中匹配的波峰。然后计算所有的匹配的波峰中，其分别对应的时间戳的时间差值，从而得到一个差值序列。该差值序列可以表示生理信号A与生理信号B之间的对应关系是否发生改变。例如，在生理信号A为心电信号、生理信号B为血氧信号的情况下，二者所对应的波峰或者波谷均是与被监测的病人的心跳所相关的，因此，计算得到的差值序列应该是一个均值序列(即该序列中的每一个元素均相等)，也就是说，在本实施例中，差值序列应为一个波动幅度较小的序列。
需要说明的是，在本实施例中，联合特征还可以包括两种不同种类的生理信号中匹配的波谷对应的时间点的差值序列，或者是，两种不同种类的生理信号的信号数据中，每一个相邻的波峰或波谷之间的周期长度的序列。
步骤S108：根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数。
在本实施例中，同源参考系数可以表示步骤S102中接收到的两种不同种类的生理信号之间是同源的可能性大小，例如，同源参考系数的取值可以是 0～100，并且同源参考系数越大，两种不同种类的生理信号是同源的可能性越大。
在一个具体的实施例中，根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数还包括：根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
根据两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列中包含的每一个元素，以及预设的同源参考系数计算步骤S102中监测到的两种不同种类的生理信号所对应的同源参考系数，例如，计算上述差值序列的平均值、或方差、均方差、最小残差，并将计算结果作为与该两种不同种类的生理信号对应的同源参考系数。
需要说明的是，在本实施例中，同源参考系数的计算方法不限于上述差值序列的平均值、或方差、均方差、最小残差等参数，也可以是其他可以直接表示两个生理信号的信号数据是否同源的参数。
步骤S110：判断所述同源参考系数是否大于预设值，若是，在执行步骤S112：确定所述两种不同种类的生理信号同源；若否，则执行步骤S114：确定所述两种不同种类的生理信号不同源。
同源参考系数是一个数值，并且，数值越大，两个不同种类的生理信号的信号数据是同源的可能性越大，反之，数值越小，两个不同种类的生理信号的信号数据来自于同一个病人的可能性越小。例如，在同源参考系数为100时，判定上述两种不同种类的生理信号是同源的，在同源参考系数为0时，判定上述两种不同种类的生理信号是不同源的。
在实际应用中，可以设置同源参考系数的阈值为80，在计算得到的同源参考系数大于80时确定两种不同种类的生理信号同源，反之，则确定两种不同种类的生理信号不同源。
在图4和图5中，给出了同源的两种不同种类的生理信号的波形匹配的示意图，在图4中，两个生理信号的信号数据中，每一个波峰之间均是对应的，且对应关系没有发生变化；在图5中，两个生理信号的有效的信号数据中，每一个波峰之间均是对应的，且对应关系没有发生变化。
在另一个实施例中，如图6所示，两个生理信号的信号数据之间，并不是 每一个波峰之间均是一一对应的，并且，其对应关系也出现了异常，在此种情况下，可明确图6中展示的两个生理信号是不同源的。
步骤S102-S114是通过计算两种不同种类的生理信号的同源参考系数来判断监测到的两种不同种类的生理信号是否是同源的，步骤S102是信号采集的步骤，步骤S104-S114是信号同源性分析的过程。
需要说明的是，在本实施例中，在接收到生理信号的信号数据之后，可选的步骤还包括，对接收到的信号数据进行预处理。具体的，预处理的方式包括了对信号数据进行滤波处理。具体的，上述接收两种不同种类的生理信号的信号数据之后还包括：对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
对信号数据进行高通滤波处理可以通过高通滤波器实现，在针对信号数据进行高通滤波处理之后，低于预设的临界值的低频信号被阻隔和减弱，也即低于预设的频率的数据会被衰减，可以用来消除生理信号的信号数据中的低频噪声。
对信号数据进行低筒滤波处理可以通过预设的低通滤波器来实现，在针对监测到的信号数据进行低通滤波之后，高于预设的临界值的高频信号会被阻隔和减弱。也就是说，高于预设频率的数据会被衰减，可以用来消除监测到的生理信号的信号数据中的高频噪声。
在本实施例中，通过对监测到的生理信号的信号数据进行高通滤波处理以及低通滤波处理，可以滤除监测到的生理信号的信号数据中的高频噪声以及低频噪声，并滤除基线漂移，从而提高信号数据的可靠性。
例如，在一个具体的实施例中，针对心电信号(EGG)进行高通滤波处理(截止频率为0.05Hz)以及低通滤波处理(截止频率为40Hz)。在另一个具体的实施例中，针对血氧信号(SPO2)进行高通滤波处理(截止频率为0.3Hz)以及低通滤波处理(截止频率为5Hz)。
上述生理信号的信号数据的采集以及分析，都是基于生理信号基于时间的变化规律或者信号特征进行分析的，相当于提取的是生理信号的信号数据在时域上的特征。在另一个实施例中，在特征提取和计算的过程中，还可以基于生理信号在频域上的特征进行。
具体的，在一个可选的实施例中，所述对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包括：对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；所述对所述两种不同种类的生理信号的信号数据进行波形匹配为：对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
也就是说，在接收到生理信号的信号数据之后，对接收到的信号数据进行傅里叶变换，进行傅里叶变换之后得到的信号数据为变换信号数据。在进行波形匹配以及特征提取等操作时，均采用进行傅里叶变换之后的变换信号数据进行操作。
需要说明的是，在本实施例中，在将信号数据从时域上的信号数据转换成频域上的信号数据时，采用的变换不止可以是傅里叶变换，还可以是任意的其他的可以实现将时域数据变换成频域数据的变换。
在另一个实施例中，还可以采用任意的变换，例如小波变换、余弦变换等变换方式，将原来接收到的生理信号的信号数据进行变化之后再进行处理。
需要说明的是，上述针对接收到的生理信号的信号数据进行傅里叶变换等变换时，可以在接收到生理信号的信号数据之后进行，也可以在对信号数据进行预处理之后进行，但是需要在对信号数据进行波形匹配之前。
在另一个可选的实施例中，针对监测到的生理信号的信号数据或者进行预处理之后的信号数据，还可以按照预设的特征提取算法，提取其对应的特征数据，然后针对特征数据进行分析。
具体的，在接收两种不同种类的生理信号的信号数据之后还包括：分别提取所述两种不同种类的生理信号的信号数据的特征数据。需要说明的是，不同种类的生理信号其对应的特征数据不一样。例如，针对心电信号(EGG)，提取其信号数据对应的特征数据的过程，可以是对信号数据进行QRS波(QRS complex)的检测和分类的过程，其中，QRS波是反映心室收缩时心脏的电行为，是心电信号自动分析的基础。再例如，针对血氧信号(SPO2)，提取其信号数据对应的特征数据的过程为监测SPO2的信号数据对应的PLUS波(脉冲波)的过程，而PLUS波是反映血氧信号的特征数据。
在提取到生理信号的信号数据的特征数据之后，即可根据该特征数据进行 特征分析，例如，将生理信号的信号数据所对应的特征数据作为步骤S102中监测到的生理信号的信号数据，以便在步骤S104-S114中进行同源性分析。
在另一个实施例中，针对单一的生理信号的信号数据所对应的特征数据，进行相应的数据分析，例如，计算与该生理信号的信号数据对应的信号质量参数，以判断检测到的信号数据是否达到预设的质量标准，如果达到，则说明监测到的信号数据达标，可以进行进一步的数据分析或者其他操作，反之，如果没有达到，则说明监测到的信号数据不达标，若用于进一步的数据分析，其所得到的数据分析结果可能因为信号数据的质量不达标从而出现较大的误差，即数据分析结果的可靠性不足。
具体的，在提取到生理信号的信号数据的特征数据之后，根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；在所述信号质量参数满足预设的信号质量参数阈值的情况下，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配。
信号质量参数(signal quality index，SQI)可以反映出监测到的生理信号的信号数据对应的信号质量的高低，并且，信号质量参数的取值范围为0～100，信号质量参数越高，其对应的生理信号的信号数据的质量越高。例如，信号数据的信噪比(signal-noise ratio，SNR)越低，信号质量参数越高。信号质量参数可以对监测到的生理信号的信号数据的质量给出客观的评价。
在一个具体的实施例中，生理信号为心电信号的情况下，其对应的特征数据为QRS波，在计算与之对应的信号参数质量时，可以先计算QRS波的信噪比(SNR)，然后通过信噪比计算与心电信号对应的信号质量参数，或者，直接以信噪比作为心电信号对应的信号质量参数。
在本实施例中，信号质量参数的计算是为了判断当前监测到的生理信号是否满足条件，具体的，监测到的生理信号的信号数据所对应的信号质量参数是否满足预设的信号质量参数阈值，若满足，则判定其满足预设的条件，是可以进行进一步的处理的。反之，若监测到的生理信号的信号数据所对应的信号质量参数不满足预设的信号质量参数阈值，则说明相应的生理信号的信号数据不 满足预设的条件，若用于进一步的数据分析，可能会产生较大的误差，因此，认为该生理信号的信号数据是无效的。
在本实施例中，针对监测到的生理信号的信号数据，不仅需要判断其是否满足一定的信号质量，还需要判断其是否存在异常或者其他情况。例如，针对心电信号对应的信号数据，还需要判断其是否存在心率异常，若存在心率异常，则需要直接进行报警提示。
具体的，分别提取所述两种不同种类的生理信号的信号数据的特征数据之后还包括：针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；判断所述单信号参数是否满足预设的单信号参数阈值；在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。针对不同种类的生理信号，其对应的信号异常或者需要进行提示的情况也会不一样，例如，针对血氧信号来讲，其需要提示的情况与心电信号的心率异常不一样，而是在血氧过低时进行提示。因此，首先需要根据生理信号的信号类型确定与该信号类型对应的单信号参数类型，例如，心电信号即为计算心率，血氧信号为计算血氧浓度。
在需要计算的单信号参数类型确定之后，即可根据预设的单信号参数计算公式来计算监测到的生理信号的信号参数所对应的单信号参数。需要说明的是，在本实施例中，在计算单信号参数时，可以是通过生理信号的信号数据来直接进行计算，也可以是通过提取到的生理信号的信号数据所对应的特征数据来计算的。
在一个具体的实施例中，生理信号的信号类型为心电信号，对应的单信号参数类型为心率，通过对心电信号的信号数据提取到的QRS波数据中QRS波的位置信息计算对应的心率。
在另一个具体的实施例中，生理信号的信号类型为血氧信号，对应的单信号参数类型为血氧浓度，通过对心电信号的信号数据提取到的PLUS波数据中PLUS波的位置信息计算对应的脉冲，并根据PLUS波的交流和直流分量比来计算与血氧信号对应的血氧浓度。
进一步的，在本实施例中，在与生理信号的信号数据对应的单信号参数计 算得到之后，还需要判断该单信号参数是否满足预设的条件，例如，判断单信号参数是否满足预设的单信号参数阈值。例如，心率超过一定值或者心率低于一定数值的情况下，均属于异常情况。在本实施例中，针对每种生理信号的信号类型，均设置有与对应的单信号参数类型对应的单信号参数阈值，用来判断计算得到的单信号参数是否满足该单信号参数阈值。若满足，则说明当前监测到的生理信号的信号数据不存在异常或者需要进行提示的情况，可以继续进行同源性分析的下一个步骤；反之，若不满足，则说明当前监测到的生理信号的信号数据存在异常或者其他需要进行提示的情况，因此，需要提示用户以便用户及时进行处理。例如，在心率过低的情况下，以报警的形式提示用户。
需要说明的是，在本实施例中，在所述单信号参数不满足预设的单信号参数阈值时，不仅需要生成相应的提示信息来提示用户，还可以根据实际情况选择是否继续进行同源性的分析。因为，在生理信号的信号数据出现的异常的情况下，某一个信号数据出现异常，而其他信号数据可能出现异常、也可能不出现异常，在此种情况下，也需要对两种不同的生理信号之间是否是同源的进行分析，以便提高对病人病情分析的可靠性。
综上所述，在本实施例中，针对接收到的生理信号的信号数据，在进行同源性分析之前，需要对接收到的生理信号的信号数据进行数据预处理、特征数据提取、计算信号质量参数、计算单信号参数等多个步骤的单个生理信号的数据分析过程。在本实施例中，上述针对接收到的单个生理信号的信号数据，对信号数据进行数据预处理、特征数据提取、计算信号质量参数、计算单信号参数的过程，即为单信号分析的过程。
如图7所示，图7展示了针对生理信号A进行单信号分析的过程。在通过监护仪监测到病人的生理信号的信号数据之后，首先对该信号数据进行预处理，以提高后续的数据分析的可靠性；然后提取信号数据的特征数据，并分别计算对应的信号质量参数以及单信号参数，并判断计算得到的信号质量参数以及单信号参数是否满足预设条件，并在均满足预设条件的情况下，继续进行下一个步骤，例如，进行波形匹配并确定波形匹配信息。
在单信号分析的过程中，单信号分析的各个步骤或者操作均是由对应的模块来完成的，例如，对生理信号A的信号数据进行预处理的过程就是由数据 预处理模块来完成的。如图8所示，图8展示了单信号分析的过程中各个模块之间的数据关系。
在本实施例中，对监护仪监测到的病人的生理信号的信号数据对病人的身体情况进行分析，例如，分析病人的病人发展趋势。不管是以何种形式的数据分析方法进行分析，都需要病人在一段时间之内的相关数据才能进行准确的分析，例如，在进行分析的数据仅为病人在30s以内的被监测到的生理信号的信号数据情况下，监测到的信号数据仅仅能够表示病人在该30s以内的病情，并且，在30s内可能受到暂时的影响，并不能完全代表病人的病情。因此，一般来讲，为了提高对监护仪监测到的生理信号的信号数据进行分析的分析结果的准确性，在本实施例中，还需要监护仪监测到的信号数据达到一定的数据量。
具体的，对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包括：针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，若是，则执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；若否，则继续执行所述接收两种不同种类的生理信号的信号数据。
例如，设置一定数据量阈值，只有在监测到的信号数据所对应的数据量达到或者超过该数据量阈值的情况下，才对监测到的数据进行分析。也就是说，只有在监测到的信号数据所对应的数据量达到或者超过预设的数据量阈值的情况下，才会进行对监测到的不同种类的生理信号的信号数据进行同源性分析，否则，继续进行生理信号的信号数据的监测，以求达到预设的数据量。
在一个具体的实施例中，可以对监测到的生理信号的信号数据进行缓冲区填充，并且，缓冲区的存储量大小是预先进行设置的。然后，只有在监测到的生理信号的信号数据对缓冲区进行填充并累积达到一定的数据量之后，才进行同源性分析，例如，在检测到的生理信号的信号数据对缓冲区的填充溢出的情况下，对监测到的不同种类的生理信号进行波形匹配或者数据预处理。
例如，如图9所示，在接收到生理信号的信号数据以及对应的特征数据、信号质量参数、单信号参数之后，即可确定当前的信号数据是否是有效的，或者确定当前的信号数据是否需要进行报警或者提示。在信号数据是有效的、且不需要进行报警或提示等特殊情况下，使用信号数据对缓冲区进行填充，并在 达到预设的数据量大小的情况下，继续进行下一个操作步骤，例如，执行步骤S104；反之，返回生理信号的信号数据的监测，即返回同源性分析的初始步骤中。
在另一个实施例中，如图10所示，图10展示了针对生理信号A和生理信号B是否同源的识别过程的流程示意图。根据图10所示，在监测到生理信号A与生理信号B的信号数据之后，首先对监测到的生理信号A的信号数据以及生理信号B的信号数据进行单信号分析(即信号预处理、特征数据提取、信号质量参数计算、单信号参数计算等)，然后进行波形匹配确定两个生理信号的信号数据之间的波形匹配信息，并根据波形匹配信息确定与之对应的联合特征，并计算生理信号A与生理信号B之间的同源参考参数，最终根据同源参考参数来判断生理信号A与生理信号B是否是同源的。进一步的，在图11中给出了上述针对生理信号A与生理信号B是否同源的识别过程中的模块的组成，即信号监测模块、单信号分析模块以及信号同源性分析模块。
在一个具体的实施例中，生理信号A为心电信号(EGG)，生理信号B为血氧信号(SPO2)，针对监护仪检测到的心电信号以及血氧信号是否是同源的进行识别的过程可以如图12所示。
需要说明的是，在本实施例中，对两种不同种类的生理信号的信号数据是否是同源的判断的过程中，不仅可以根据同源参考系数来判断二者是否是同源的，还可以根据上述两桶不同种类的生理信号的信号数据的其他特征数据来进行判断。
在一个具体的实施例中，如图13所示，除了获取生理信号的信号数据所对应的特征数据、信号质量参数、单信号单数、联合特征以及同源参考系数之外，还可以获取生理信号A和生理信号B间期差(同一个生理信号下两个波形变化周期长度之间的差值)、峰峰差均值(生理信号A的波峰与生理信号B的波峰所对应的时间戳之间的差值的平均值)、最大值(生理信号A的波峰与生理信号B的波峰所对应的时间戳之间的差值的最大值)、最小值(生理信号A的波峰与生理信号B的波峰所对应的时间戳之间的差值的最小值)和比例。需要说明的，在判断两个不同种类的生理信号是否是同源的判断过程中，获取的参数不限于上述列举的参数，还可以包括其他可以展示两个信号数据之间的 同源性关系的其他参数或者特征数据。
如图13所示，根据生理信号A与生理信号B之间的若干个峰峰差的具体值是否保持不变、峰峰差是否小于间期的预设比例大小、最大值与最小值之间的差距等特征，逐步判断生理信号A的波形数据与生理信号B的波形数据是否是同步的，在绝对同步的情况下直接将同源参考系数设置为100，即生理信号A与生理信号B是同源的，在绝对不同步的情况下直接将同源参考系数设置为0，即生理信号A与生理信号B是不同源的。在不完全同步的情况下，根据峰峰差的变化规律，计算生理信号A的波峰数据与生理信号B的波峰数据之间同步的比例数，并根据该比例数以及同源性判断的历史数据来确定生理信号A与生理信号B是否是同源的。
此外，为解决传统技术中的对监护仪监测到的多种生理信号进行多参数融合分析时因为不能区别当前的多种生理信号是否来自于同一个病人而存在的可信度不足的技术问题，在一个实施例中，如图14所示，还提出了一种生理信号的同源性识别装置，包括信号数据接收模块102、波形匹配模块104、联合特征获取模块106、同源参考系数计算模块108、同源识别模块110，其中：
信号数据接收模块102，用于接收两种不同种类的生理信号的信号数据；
波形匹配模块104，用于对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
联合特征获取模块106，用于根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
同源参考系数计算模块108，用于根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
同源识别模块110，用于在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
可选的，在其中一个实施例中，联合特征获取模块106还用于根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。
可选的，如图14所示，在其中一个实施例中，上述装置还包括数据预处理模块112，用于对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
可选的，如图14所示，在其中一个实施例中，上述装置还包括信号质量参数计算模块114，用于分别提取所述两种不同种类的生理信号的信号数据的特征数据；根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；在所述信号质量参数满足预设的信号质量参数阈值的情况下，调用所述波形匹配模块104。
可选的，如图14所示，在其中一个实施例中，上述装置还包括单信号参数计算模块116，用于针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；判断所述单信号参数是否满足预设的单信号参数阈值；在所述单信号参数满足预设的单信号参数阈值时，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
可选的，如图14所示，在其中一个实施例中，上述装置还包括缓冲区填充模块118，用于针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，在所述信号数据的数据量大小大于或等于预设的数据量阈值的情况下，调用所述波形匹配模块104；在所述信号数据的数据量大小小于预设的数据量阈值的情况下，调用所述信号数据接收模块102。
可选的，在其中一个实施例中，同源参考系数计算模块108还用于根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在其中一个实施例中，同源参考系数计算模块108还用于计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
可选的，如图14所示，在其中一个实施例中，上述装置还包括信号数据变换模块120，用于对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；所述波形匹配模块还用于对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
实施本发明实施例，将具有如下有益效果：
采用了上述生理信号的同源性识别方法和装置之后，在监护仪监测到多种不同种类的生理信号的信号数据的情况下，通过对监测到的不同种类的生理信号的信号数据的联合特征的提取和计算，来确定可以标识两种不同种类的生理信号是同源的可能性的同源参考参数的大小，从而判断两个生理信号是否是同源的。也就是说，可以对监护仪监测到的多项数据自动进行同源性分析，避免了监护仪在同时监测不同的病人的生理数据的情况下对监测到的多项数据进行融合分析的过程中，因为无法区别监测到的数据是否是来自于同一个病人导致的分析结果不准确的缺陷，从而提高了数据分析的结果的准确度和可信度。
在上述实施例中，可以全部或部分的通过软件、硬件、固件或者其任意组合来实现。当使用软件程序实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本发明实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或者数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或者数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或半导体介质(例如固态硬盘Solid State Disk(SSD))等。
在一个实施例中，如图15所示，图15展示了一种运行上述生理信号的同源性识别方法的基于冯诺依曼体系的计算机系统的终端。该计算机系统可以是智能手机、平板电脑、掌上电脑、笔记本电脑或个人电脑等终端设备。具体的，可包括通过系统总线连接的外部输入接口1001、处理器1002、存储器1003和输出接口1004。其中，外部输入接口1001可选的可至少包括网络接口10012。存储器1003可包括外存储器10032(例如硬盘、光盘或软盘等)和内存储器10034。输出接口1004可至少包括显示屏10042等设备。
在本实施例中，本方法的运行基于计算机程序，该计算机程序的程序文件存储于前述基于冯诺依曼体系的计算机系统的外存储器10032中，在运行时被加载到内存储器10034中，然后被编译为机器码之后传递至处理器1002中执行，从而使得基于冯诺依曼体系的计算机系统中形成逻辑上的信号数据接收模块102、波形匹配模块104、联合特征获取模块106、同源参考系数计算模块108、同源识别模块110、数据预处理模块112、信号质量参数计算模块114、单信号参数计算模块116以及缓冲区填充模块118。且在上述生理信号的同源性识别方法执行过程中，输入的参数均通过外部输入接口1001接收，并传递至存储器1003中缓存，然后输入到处理器1002中进行处理，处理的结果数据或缓存于存储器1003中进行后续地处理，或被传递至输出接口1004进行输出。
具体的，处理器1002用于执行如下操作：
接收两种不同种类的生理信号的信号数据；
对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
可选的，在一个实施例中，处理器1002还用于根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理 信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。
可选的，在一个实施例中，处理器1002还用于执行对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
可选的，在一个实施例中，处理器1002还用于执行分别提取所述两种不同种类的生理信号的信号数据的特征数据；根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；在所述信号质量参数满足预设的信号质量参数阈值的情况下，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配。
可选的，在一个实施例中，处理器1002还用于执行针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；判断所述单信号参数是否满足预设的单信号参数阈值；在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
可选的，在一个实施例中，处理器1002还用于执行针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，若是，则执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；若否，则继续执行所述接收两种不同种类的生理信号的信号数据。
可选的，在一个实施例中，处理器1002还用于执行根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在一个实施例中，处理器1002还用于执行计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
可选的，在一个实施例中，处理器1002还用于对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
以上所揭露的仅为本发明较佳实施例而已，当然不能以此来限定本发明之权利范围，因此依本发明权利要求所作的等同变化，仍属本发明所涵盖的范围。

Claims (20)

  1. 一种生理信号的同源性识别方法，其特征在于，包括：
    接收两种不同种类的生理信号的信号数据；
    对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
    根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
    根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
    在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
  2. 根据权利要求1所述的生理信号的同源性识别方法，其特征在于，所述根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征包括：
    根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。
  3. 根据权利要求1所述的生理信号的同源性识别方法，其特征在于，所述接收两种不同种类的生理信号的信号数据之后还包括：
    对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
  4. 根据权利要求1所述的生理信号的同源性识别方法，其特征在于，所述接收两种不同种类的生理信号的信号数据之后还包括：
    分别提取所述两种不同种类的生理信号的信号数据的特征数据；
    根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；
    在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无效的；
    在所述信号质量参数满足预设的信号质量参数阈值的情况下，执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配的步骤。
  5. 根据权利要求4所述的生理信号的同源性识别方法，其特征在于，所述分别提取所述两种不同种类的生理信号的信号数据的特征数据之后还包括：
    针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；
    根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；
    判断所述单信号参数是否满足预设的单信号参数阈值；
    在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
  6. 根据权利要求1所述的生理信号的同源性识别方法，其特征在于，所述对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包括：
    针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，若是，则执行所述对所述两种不同种类的生理信号的信号数据进行波形匹配；若否，则继续执行所述接收两种不同种类的生理信号的信号数据。
  7. 根据权利要求1所述的生理信号的同源性识别方法，其特征在于，所述根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数还包括：
    根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
  8. 根据权利要求7所述的生理信号的同源性识别方法，其特征在于，所述根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数还包括：
    计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
  9. 根据权利要求1至8任一所述的生理信号的同源性识别方法，其特征在于，所述对所述两种不同种类的生理信号的信号数据进行波形匹配之前还包 括：
    对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；
    所述对所述两种不同种类的生理信号的信号数据进行波形匹配具体为：
    对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
  10. 一种生理信号的同源性识别装置，其特征在于，所述
    信号数据接收模块，用于接收两种不同种类的生理信号的信号数据；
    波形匹配模块，用于对所述两种不同种类的生理信号的信号数据进行波形匹配，确定所述两种生理信号的信号数据中的波形匹配信息；
    联合特征获取模块，用于根据所述波形匹配信息计算所述两种不同种类的生理信号的信号数据之间的联合特征；
    同源参考系数计算模块，用于根据所述联合特征计算与所述两种不同种类的生理信号对应的同源参考系数；
    同源识别模块，用于在所述同源参考系数大于预设值的情况下，确定所述两种不同种类的生理信号同源。
  11. 根据权利要求10所述的生理信号的同源性识别装置，其特征在于，所述联合特征获取模块还用于根据所述波形匹配信息确定所述两种不同种类的生理信号中匹配的波峰，计算所述两种不同种类的生理信号中匹配的波峰对应的时间点的差值序列，将所述差值序列作为所述两种不同种类的生理信号的信号数据之间的联合特征。
  12. 根据权利要求10所述的生理信号的同源性识别装置，其特征在于，所述装置还包括数据预处理模块，用于对所述两种不同种类的生理信号的信号数据进行高通滤波处理和低通滤波处理。
  13. 根据权利要求10所述的生理信号的同源性识别装置，其特征在于，所述装置还包括信号质量参数计算模块，用于分别提取所述两种不同种类的生理信号的信号数据的特征数据；根据预设的信号质量参数计算公式计算与所述特征数据对应的信号质量参数；在所述信号质量参数不满足预设的信号质量参数阈值的情况下，确定与该信号质量参数对应的所述生理信号的信号数据是无 效的；在所述信号质量参数满足预设的信号质量参数阈值的情况下，调用所述波形匹配模块。
  14. 根据权利要求13所述的生理信号的同源性识别装置，其特征在于，所述装置还包括单信号参数计算模块，用于针对任意种类的生理信号，确定该生理信号的信号类型，确定与所述信号类型对应的单信号参数类型；根据预设的单信号参数计算公式以及所述特征数据，计算与该生理信号对应的单信号参数；判断所述单信号参数是否满足预设的单信号参数阈值；在所述单信号参数不满足预设的单信号参数阈值时，生成信号异常的提示信息并提示用户。
  15. 根据权利要求10所述的生理信号的同源性识别装置，其特征在于，所述装置还包括缓冲区填充模块，用于针对所述两种不同种类的生理信号的信号数据，判断所述信号数据的数据量大小是否大于或等于预设的数据量阈值，在所述信号数据的数据量大小大于或等于预设的数据量阈值的情况下，调用所述波形匹配模块；在所述信号数据的数据量大小小于预设的数据量阈值的情况下，调用所述信号数据接收模块。
  16. 根据权利要求10所述的生理信号的同源性识别装置，其特征在于，所述同源参考系数计算模块还用于根据预设的同源参考系数计算公式以及所述差值序列，计算与所述两种不同种类的生理信号对应的同源参考系数。
  17. 根据权利要求16所述的生理信号的同源性识别装置，其特征在于，所述同源参考系数计算模块还用于计算所述差值序列的平均值或均方差作为与所述两种不同种类的生理信号对应的同源参考系数。
  18. 根据权利要求10-17任一所述的生理信号的同源性识别装置，其特征在于，所述装置还包括信号数据变换模块，用于对所述两种不同种类的生理信号的信号数据分别进行傅里叶变换，得到变换信号数据；
    所述波形匹配模块还用于对所述两种不同种类的生理信号的信号数据对应的变换信号数据进行波形匹配。
  19. 一种计算机可读存储介质，包括计算机指令，当所述计算机指令在计算机上运行时，使得计算机执行如权利要求1-9所述的方法。
  20. 一种生理信号的同源性识别终端，其特征在于，包括：
    处理器和存储器；
    其中，所述处理器通过调用所述存储器中的代码或指令以执行如权利要求1至9任意一项所述的方法。
PCT/CN2017/110350 2017-05-03 2017-11-10 生理信号的同源性识别方法及装置 WO2018201683A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201780086365.2A CN110312466A (zh) 2017-05-03 2017-11-10 生理信号的同源性识别方法及装置
US16/672,640 US11457873B2 (en) 2017-05-03 2019-11-04 Method and apparatus for identifying homology of physiological signals

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710304322.9A CN108784650A (zh) 2017-05-03 2017-05-03 生理信号的同源性识别方法及装置
CN201710304322.9 2017-05-03

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US16/672,640 Continuation-In-Part US11457873B2 (en) 2017-05-03 2019-11-04 Method and apparatus for identifying homology of physiological signals
US16/672,640 Continuation US11457873B2 (en) 2017-05-03 2019-11-04 Method and apparatus for identifying homology of physiological signals

Publications (1)

Publication Number Publication Date
WO2018201683A1 true WO2018201683A1 (zh) 2018-11-08

Family

ID=64015937

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/110350 WO2018201683A1 (zh) 2017-05-03 2017-11-10 生理信号的同源性识别方法及装置

Country Status (3)

Country Link
US (1) US11457873B2 (zh)
CN (2) CN108784650A (zh)
WO (1) WO2018201683A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108784650A (zh) * 2017-05-03 2018-11-13 深圳迈瑞生物医疗电子股份有限公司 生理信号的同源性识别方法及装置
CN110558961B (zh) * 2019-09-10 2022-03-22 哈尔滨亿尚医疗科技有限公司 生理信息显示方法、设备和计算机可读存储介质
CN113662510B (zh) * 2020-05-15 2023-03-10 华为技术有限公司 基于多信号的用户健康状态检测方法、可穿戴设备和介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5719950A (en) * 1994-03-24 1998-02-17 Minnesota Mining And Manufacturing Company Biometric, personal authentication system
CN1931091A (zh) * 2006-10-12 2007-03-21 北京飞天诚信科技有限公司 一种身份鉴定方法及系统
CN101773394A (zh) * 2010-01-06 2010-07-14 中国航天员科研训练中心 身份识别方法及应用该方法的身份识别系统
CN102231213A (zh) * 2011-06-29 2011-11-02 哈尔滨工业大学深圳研究生院 Ecg门禁卡身份识别方法及系统
CN104054038A (zh) * 2011-10-26 2014-09-17 索尼电脑娱乐公司 个体鉴别装置和个体鉴别方法
CN105740680A (zh) * 2014-12-30 2016-07-06 三星电子株式会社 用于基于生物信号对用户进行认证的方法和设备
CN105787420A (zh) * 2014-12-24 2016-07-20 北京三星通信技术研究有限公司 用于生物认证的方法、装置以及生物认证系统

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
MX9702434A (es) * 1991-03-07 1998-05-31 Masimo Corp Aparato de procesamiento de señales.
AU2002322302A1 (en) * 2001-06-25 2003-01-08 Science Applications International Corporation Identification by analysis of physiometric variation
US6773397B2 (en) * 2001-10-11 2004-08-10 Draeger Medical Systems, Inc. System for processing signal data representing physiological parameters
EP1525710A4 (en) * 2002-07-29 2008-02-13 Idesia Ltd METHOD AND APPARATUS FOR RECOGNIZING ELECTROBIOMETRIC IDENTITY
US8594773B2 (en) * 2007-04-25 2013-11-26 Siemens Medical Solutions Usa, Inc. Denoising and artifact rejection for cardiac signal in a sensis system
CN101933811B (zh) * 2010-09-03 2016-06-22 深圳市索莱瑞医疗技术有限公司 一种血氧饱和度检测方法及系统
WO2012151680A1 (en) * 2011-05-10 2012-11-15 Agrafioti Foteini System and method for enabling continuous or instantaneous identity recognition based on physiological biometric signals
US20150065830A1 (en) * 2013-08-30 2015-03-05 General Electric Company System and method for operating a sensor for determining blood characteristics of a subject
CN103690152B (zh) * 2014-01-06 2016-04-20 山东大学 一种基于脉搏解析的动脉弹性功能评估装置
US10602944B2 (en) * 2014-05-08 2020-03-31 Drägerwerk AG & Co. KGaA Detecting artifacts in a signal
US10154818B2 (en) * 2014-12-24 2018-12-18 Samsung Electronics Co., Ltd. Biometric authentication method and apparatus
CN104939820B (zh) * 2015-05-28 2017-12-19 深圳市理邦精密仪器股份有限公司 一种起搏信号检测方法及装置
CN108784650A (zh) * 2017-05-03 2018-11-13 深圳迈瑞生物医疗电子股份有限公司 生理信号的同源性识别方法及装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5719950A (en) * 1994-03-24 1998-02-17 Minnesota Mining And Manufacturing Company Biometric, personal authentication system
CN1931091A (zh) * 2006-10-12 2007-03-21 北京飞天诚信科技有限公司 一种身份鉴定方法及系统
CN101773394A (zh) * 2010-01-06 2010-07-14 中国航天员科研训练中心 身份识别方法及应用该方法的身份识别系统
CN102231213A (zh) * 2011-06-29 2011-11-02 哈尔滨工业大学深圳研究生院 Ecg门禁卡身份识别方法及系统
CN104054038A (zh) * 2011-10-26 2014-09-17 索尼电脑娱乐公司 个体鉴别装置和个体鉴别方法
CN105787420A (zh) * 2014-12-24 2016-07-20 北京三星通信技术研究有限公司 用于生物认证的方法、装置以及生物认证系统
CN105740680A (zh) * 2014-12-30 2016-07-06 三星电子株式会社 用于基于生物信号对用户进行认证的方法和设备

Also Published As

Publication number Publication date
US20200060625A1 (en) 2020-02-27
US11457873B2 (en) 2022-10-04
CN110312466A (zh) 2019-10-08
CN108784650A (zh) 2018-11-13

Similar Documents

Publication Publication Date Title
US10709392B2 (en) Sensor fault detection using analyte sensor data pattern comparison
US9042973B2 (en) Apparatus and method for measuring physiological signal quality
US11457873B2 (en) Method and apparatus for identifying homology of physiological signals
WO2011150585A1 (zh) 生理信号质量评估系统及方法
CN107788969A (zh) 一种心电信号中qrs波群的自动检测方法
Paula Rocha et al. Beyond long memory in heart rate variability: an approach based on fractionally integrated autoregressive moving average time series models with conditional heteroscedasticity
US20200352504A1 (en) Image Drunken Driving Judgment System and Related Method
CN107276779B (zh) 一种监控方法、系统及设备
WO2023226223A1 (zh) Ppg信号质量评估方法及装置以及ppg信号处理方法及系统
CN111839494A (zh) 一种心率监测方法及系统
US20220351384A1 (en) Method, apparatus and program
US20210000384A1 (en) Method and apparatus for monitoring a human or animal subject
US20080281216A1 (en) Cardiac Arrhythmias Analysis of Electrophysiological Signals Based on Symbolic Dynamics
CN111696664B (zh) 一种心衰疾病检测系统
CN114145738A (zh) 分析物浓度数据生成方法和装置、监测分析物水平的系统
JPH11155845A (ja) ストレス度判定装置
WO2023108331A1 (zh) 自适应实时心电信号质量评估方法
Ibrahimy et al. An emergency medical care network system for fetal ECG monitoring
CN114271799B (zh) 一种基于智能终端的心血管检测方法
CN116763312B (zh) 一种基于可穿戴设备的异常情绪识别方法及系统
CN117292843B (zh) 电信号数据处理方法、装置、设备和存储介质
CN117954116B (zh) 基于人工智能的呼吸重症患者状态监测预警方法
US20240180495A1 (en) Systems and methods for assessing ecg lead signal quality
EP4368115A1 (en) Cardiotocographic scanning session duration
CN113974554A (zh) 重搏波识别方法、装置、设备和计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17908510

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17908510

Country of ref document: EP

Kind code of ref document: A1